Least Squares Parameter Es.ma.on

Size: px
Start display at page:

Download "Least Squares Parameter Es.ma.on"

Transcription

1 Least Squares Parameter Es.ma.on Alun L. Lloyd Department of Mathema.cs Biomathema.cs Graduate Program North Carolina State University

2 Aims of this Lecture 1. Model fifng using least squares 2. Quan.fica.on of uncertainty in parameter es.mates: asympto.c sta.s.cal theory 3. Sensi.vity analysis (local linear analysis)

3 Introduc.on SIR model: Mathema.cal model for the spread of an infec.on through a popula.on by person- to- person contact This mathema.cal model describes the.me evolu.on of the state variables S, I and R, using a set of differen.al equa.ons, whose terms describe the rates at which various biological processes (infec.on and recovery) happen ds/ = βsi/n S infection I recovery R di/ = βsi/n γ I dr/ = γ I. Parameters: constants that appear in the func.ons describing these rates

4 The forward problem S infection I recovery R ds/ = βsi/n di/ = βsi/n γ I dr/ = γ I. If we know the model equa.ons, the values of the parameters and the ini.al values of the states, we can (numerically) solve the model forwards in.me to find out how the state variables will evolve Forward simula.on of S and I, given β, γ, N, S() and I() : Forward problem: Parameters dynamical behavior

5 The inverse problem (parameter es.ma.on) In many cases, we don t know the values of the parameters, but we have observa.ons of the system (e.g..me series data) Confined to bed Convalescent.-----o 25 U, w 2 a, 15 n E Z January 1. '26 ' 28' - l '3 - Februo3ry Can we figure out parameter values that would make the model exhibit behavior that is consistent with observed data? Inverse problem: dynamical behavior parameters

6 Least- Squares FiFng One approach to find unknown parameter values: fit a model to.me series data trajectory matching : find parameter values for which the model provides the best fit to a data set What do we mean by best fit? Least squares criterion: (ordinary least squares) Minimize sum of squared differences between model predic9ons and observed data - sum of squared errors hopefully familiar from simple linear regression n ( ( ) ) 2 F(θ) = y model predicted ( t i ;θ) y observed t i i=1 Here, θ is the vector of parameters

7 Least- Squares FiFng y Least squares criterion: minimize sum of squared differences between model predic.ons and observed data (sum of squared errors) y Model Predicted (t 1 ; θ) e 1 e 2 (t 1,y 1 observed) (t 2,y 2 observed) e 3 (t 3,y 3 observed) n F(θ) = y model predicted ( t i ;θ) y observed t i i=1 ( ( ) ) 2 Dots: data points (observa.ons) Red crosses: model predic.ons (for some value of the parameter vector) t e 4 (t 4,y 4 observed) Green lines indicate (model- data): called errors or residuals [Why do we use sum of squares? 1. Cannot simply add errors (cancela.on of + and - errors) 2. Sum of squared errors is beeer behaved than summed absolute errors.]

8 Minimiza.on problem Least- Squares FiFng ˆ θ = # % argmin { θ feasible} % & i=1 n ' ( y model predicted ( t i ;θ ) y observed ( t i ) ) 2 % ( % ) 1. Cannot solve this analy.cally in general (linear regression is an excep.on), so will probably need to do op.miza.on numerically Process can be non- trivial, e.g. if the func.on F(θ) has mul.ple local minima 2. If our model is an ODE, we probably don t have a formula for the model s output, so will have to numerically solve the ODE to make predic.ons If the ini.al condi.on of our ODE is unknown/uncertain, we can include it in the list of quan..es to be es.mated it 3. Es.mates depend on the data: would get a different es.mate if data was different Mathema.cally: the best- fifng parameter vector is a random variable, and its es.ma.on is a sta.s.cal process What are the uncertain.es in the es.mates we obtain? standard errors for parameter es.mates?

9 More Generally... State Space Nota.on Model: dx = f(x, t; ), where x and f are m dimensional vectors, and θ is a p dimensional vector of parameters Observa.on func.on (maps model to observable quan.ty): Ordinary least squares parameter es.mate is ˆ θ = y(t) =h(x, t; ) # % argmin { θ feasible} % & i=1 n ' ( y model predicted ( t i ;θ) y observed ( t i ) ) 2 % ( % )

10 Uncertainty Es.mates: Intui.on Nonlinear regression theory can be used to provide uncertainty es.mates Intui.on: consider two graphs of error sum of squares, based on two different models.8.7 Error Sum of Squares Parameter value Which gives us more certainty in the parameter es.mate? For 2 there is a larger range of parameter values that gives a reasonable fit In some sense, there is less informa.on about the true value of the parameter Intui.on: the curvature (2 nd deriva.ve) of the SS func.on at its minimum is an inverse measure of the informa.on Small second deriva.ve: less informa.on/more uncertainty in value Idea is formalized in the no.on of Fisher informa.on matrix (coming soon)

11 Uncertainty Es.mates: Intui.on.8.7 A couple of comments: Error Sum of Squares Parameter value By the chain rule, deriva.ves of the sum of squares func.on F ( ) = X i y predicted (t i ; ) y observed (t i ) 2 with respect to the parameters will involve deriva.ves of model predic.ons with respect to the parameters (we call these sensi.vi.es) It will be difficult to es9mate insensi9ve parameters 2. Model 2 appears to be a beeer fit (smaller value for error sum of squares) Is there a meaningful no.on of how improved the fit is?

12 Uncertainty Es.mates We need a sta.s.cal model for our observa.ons describes the origin of the errors in the data Simplest sta.s.cal model: observa.ons are the values predicted by the model under the true parameter values plus observa.on error y observed ( t i ) = y model predicted ( t i ;θ true ) + e i e i are the observa.on errors In the simplest sefng, we assume errors are independent, iden.cally distributed and have constant variance, σ 2 note: following theory is exact if errors are normal, but we don t need to assume errors are normal if we appeal to large sample size theory (asympto.c theory) can also view the following as providing a bound on uncertainty

13 We ask the ques.on: Uncertainty Es.mates Suppose we could watch the system evolve over.me on mul.ple occasions, giving us a collec.on of data sets [because we have different realiza.ons of the noise process], and es.mated the parameter values for each of those data sets... How much varia.on would we see in the parameter es.mates? Could characterize this varia.on by the variance in the values of the es.mates Could also examine the correla.on between es.mates of different parameters, captured by the covariance between their values All this informa.on is summarized in the covariance (variance/covariance) matrix, Σ Var( ˆ) cov(ˆ, ˆ) For SIR example: = cov( ˆ, ˆ) Var(ˆ)

14 Uncertainty Es.mates for Parameters Asympto.c (large sample size) theory says that the parameter es.mator (a random variable) has a mul.variate normal distribu.on, centered on the true parameter vector and with variance- covariance matrix = 2 ( ) T ( ) Here χ is the n x p matrix of sensi.vi.es, with entries (discuss calcula.on of sensi.vi.es soon!) # of data points # of parameters es.mated ( )= 1 ( ) ij (t i ; 1 ; 1 ; 2 ; 2 ; n ; n ; 1 ; 2 ; n ; p 1 C A Problem: we don t know θ or σ 2 (true values), so we make use of our es.mates

15 <latexit sha1_base64="bpgprwa82fidlhwlnyk5qfxpr1c=">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</latexit> sha1_base64="bpgprwa82fidlhwlnyk5qfxpr1c=">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</latexit> sha1_base64="srv+yxncxqurtdrjatzg1varh1m=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> sha1_base64="kshwnz+gkgbcovxit26el/z5lm=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> <latexit sha1_base64="feyzcsbqj61lrrcowietri1dsp8=">aaab5hicbvdlsgnbeoz1gddx9oplmaiewq4xpqpepeywdhcmj3ttybmzi4zvujy8gmevih49zu8+tdohqamfjquvd1dw5kpac4mvb2nza3tmt7pn7b/7hxh1pgwzwghsikxlphnxipqbjikhz3cie8jhe1ofdfz29ormzi1y7kd8qguibscnnqbvwlap5mdrjfysgiwxqh724kwukwosilvbdyoc+iu3jixcqd8rlozcjpkqu45qnqltl/mzp+zcktflmunke5urvydknlo7sspxmxia2vvvjv7ndqtkbvql1hlbqmviuviorhmb/cxiavcqmjjchzhuvizg3hbblhnfzrcufrxowlf1mkihdz9hqaxo4bwuiyrruiv7aeatbmtwdk/eyhvx3rz3reogt5w4ht/wpr4bgdqlag==</latexit> <latexit sha1_base64="iqio66ykhuvosw9lnpzfddkzvtk=">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</latexit> <latexit sha1_base64="iqio66ykhuvosw9lnpzfddkzvtk=">aaacsxicbvbna9wwfjq3/ui3h9mmx1xeq2edybf7aaaucrnkmei3caw27rp8hcurzsm9fxbjv5dlb73lp+ssqlokfluo+ka4jhzt6tnemllamwvah6k48ep3m6+qz//mxlv2ud1+shrqytxlesdwmpencolcexkdj4vfmeitf4mjztdv7hd7roleyrzsqcfnbivkykkjfiwtchc3xcdm1woxsui8fw3cixfdp+epkf2ygklcnqfhbccftwvz1vkjbthxsrj2jb5edlyfz3eihep2/hmygo3ao/pbes7ljltipbz9fwsq6qensg3otkkxo2nr7pca2l2qhfcgzomgjpwykdnnm3ktl33kl5vlp/the5+rfewuzs2kxcclonzd9zrxf96kpmxn2iht1yrgli7kas2p5f2tpfuwjemzjyct8m/lmgdfipnyuxki+19+sa7ej6jwfh35wzbzrvslruyih1gn9ke22djjtk5u2tx7cb4evwft8gvrbqxlgfesh/qw7kdscszvq==</latexit> <latexit sha1_base64="fvnr7f4hpmwdfd6gseskeqv6qg=">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</latexit> <latexit sha1_base64="kshwnz+gkgbcovxit26el/z5lm=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">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</latexit> <latexit sha1_base64="1bewaplawhwrmtko43hwjhpmdjo=">aaacsxicbvdpa9ragj1sq9b116phlmxyqtlsasgigkhf48vum1hzxu/tl44myealsit8e168eev/4eepinjqzhcptfxbwoo993z85jkkdhebh1tbv3l23cb//4ogjx8gt58durk2eiey1ku9tschvgynpejjcwuriktjuxk+1/lhn9e6vzodmlc4k+duqexjic/fg9c5uqkgzmtdiqor4ktuffi+6x9jzilshevwfkg+fykebbgfnuqjghajihwb8w2yifwk+1veinez2/hgzdcbgav2ifrm+/8ew2i8h3rayrpaq1kdc9morgjwdhulxryvaocvyhm4xamnbgps2brrmtfeixlwwn9mcqx6vwjbgrn5kxikwvq7m56nfg/b1pt9mbwkfpvheyul8pqzankxa8vryl6bknik3yb+uyb18g+fk7eqkbx75ndl+no3acfdwz7u4s22ab7axbzcmwsddsl31g+2zcjpvcvrnf7hfwnfgz/an+lqo9ydxznp2d3tolb5q/a==</latexit> Uncertainty Es.mates for Parameters Asympto.c (large sample size) theory says that the parameter variance- covariance matrix for the es.mated parameters is Here, ˆ2 (n) (ˆ ) ˆ =ˆ2 (n) (ˆ ) T 1 (n) (ˆ ) is our least squares es.mate of the parameters is (minimized value of error sum of squares) / (n p) n data points (13 for our SIR example), p es.mated parameters (2 for SIR) is the n x p matrix of sensi.vi.es, with entries (n) (ˆ ) 1 ; ˆ 2 ; ˆ n ; ˆ 1 Unbiased es.mate (n) ( ) ij (t i ; ˆ 1 ; ˆ 2 ; ˆ n ; ˆ 1 ; ˆ 2 ; ˆ n ; ˆ p 1 C A

16 <latexit sha1_base64="wsksbqs8ylpqupn84l1gfncwci4=">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</latexit> <latexit sha1_base64="4kbbco1vksaolv/dftjkiiejgxu=">aaacs3icbvbnswmxfmxwq7v+vt16crahhvj2paaxosccx4q2fbqlznosg81ulurtosz9f168epnpepggiaftjnboxcyzmzjjepfgmuw7xcrs7ae3djmbew3d3b39gshhytevzkoh1ynhnbm8yk3ginhdrbgjpcha3vpv2g8pmnjcrvcwjfk3ji8r9zklykrewxnvipeldn1fajq6ksrudkylnydgqsca9hhl7vjuhk1td9elobjbwawbfk9qtkv2bpg/cwakigzo9apvbl/sjgqrueg7jh2dn2ukobusfhettslcxmj6xjaercprvppisrpjvkh/tsmrmbnqjzeykjtr6gnkmgbak97i3fvv4naf+im/iotobfdlritwqgicff4j5xjiiygkko4uatmabetaqm/rwpwvn+8n/soqs6w5rrxrtvkdoxsmtlajoegc1denaqamougffaav9g29wp/wj/u7jwas2cwrwkam+wffrpi</latexit> <latexit sha1_base64="wsksbqs8ylpqupn84l1gfncwci4=">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</latexit> <latexit sha1_base64="wsksbqs8ylpqupn84l1gfncwci4=">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</latexit> <latexit sha1_base64="wsksbqs8ylpqupn84l1gfncwci4=">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</latexit> Uncertainty Es.mates for Parameters Standard errors (SE: standard devia.ons for the es.mated parameters) can be calculated by taking square root of appropriate entry on diagonal entry of Σ No.ce that es.mates of different parameters will typically be correlated Calculate correla.on between parameter es.mates using = cov(ˆ i, ˆ j ) SE(ˆ i )SE(ˆ j )

17 <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="ehhzjncyzpsom7wi2yxooqx14gi=">aaacd3icbzdlsgmxfiyz9vbrrerstbaodvnmpkdlghvdvbax6azltjppq5ozickizzg3cooruhghifu37nwbm21bbfh8pgfc5kc3485u9q2v6zcyura+kzxs7s1vbo7v94/aksokys2smqj2fvbuc5c2tjmc9qnjqxhc9rxx1d5vxnppwjrekcnmfuedemwmalawp3yqrtiikkbg9qmol7jftgdghcqvfvzqv+u2dv7krwmzhwqak5mv/zpdikscbpqwkgpnmph2kvzmwmnwclnfi2bjgfiewzdefr56xsfdj8yz4cdsjotajx1f+kijsacn9ctajtvjlzf9qvuqhl17kwjjrncszh4keyx3hpbw8yjiszscggehm/orjcexa2ksyh+asrrwm7foay9ec23qluz/huurh6bhvkymuuandoyzqiyiebn6qa/wo/vsvvnvs9acnz85rh9kfxwdmygcsg==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="feyzcsbqj61lrrcowietri1dsp8=">aaab5hicbvdlsgnbeoz1gddx9oplmaiewq4xpqpepeywdhcmj3ttybmzi4zvujy8gmevih49zu8+tdohqamfjquvd1dw5kpac4mvb2nza3tmt7pn7b/7hxh1pgwzwghsikxlphnxipqbjikhz3cie8jhe1ofdfz29ormzi1y7kd8qguibscnnqbvwlap5mdrjfysgiwxqh724kwukwosilvbdyoc+iu3jixcqd8rlozcjpkqu45qnqltl/mzp+zcktflmunke5urvydknlo7sspxmxia2vvvjv7ndqtkbvql1hlbqmviuviorhmb/cxiavcqmjjchzhuvizg3hbblhnfzrcufrxowlf1mkihdz9hqaxo4bwuiyrruiv7aeatbmtwdk/eyhvx3rz3reogt5w4ht/wpr4bgdqlag==</latexit> <latexit sha1_base64="w+ad7x6uub+qljszdovdo1ozwla=">aaacd3icbzc9tsmwfivvyn/5kzcywcbqwaqebuykfthaoqvse1u3rtnatzpidpcqkg/awquwmiaqkysbb4ptvgikr7l6dx7bd8tpojr47qftmvufmfxaxmlurq2vrfz29pu6srtldvpihlvdlezwwpwnnwi1k4vqxkkdhsoz8v67r1tmifxjrmlljdyj3nekrprdwuhfqsq5n6kynau5ll4zr+pumjrnfvbm3fbbhjkb/gtwefprrq1j78xkizywjdbwrd8dzubhl5mxwsqpqzzinsifzzx2kmkukgh+9tkapr9eiukhtiq8buz4kcpdyjgdpoiwagz2ul+v+tk5nonmh5ngagxxtyujqjyhjshkn6xdfqxmgcusxtxwkdoa3i2ajlelzzlf9c67jhuq3v+jsnwizd2im6ehacz3abv9aecvfwcm/w4jw4t86r8zzprtjtmr34jef9c3udm9i=</latexit> <latexit sha1_base64="w+ad7x6uub+qljszdovdo1ozwla=">aaacd3icbzc9tsmwfivvyn/5kzcywcbqwaqebuykfthaoqvse1u3rtnatzpidpcqkg/awquwmiaqkysbb4ptvgikr7l6dx7bd8tpojr47qftmvufmfxaxmlurq2vrfz29pu6srtldvpihlvdlezwwpwnnwi1k4vqxkkdhsoz8v67r1tmifxjrmlljdyj3nekrprdwuhfqsq5n6kynau5ll4zr+pumjrnfvbm3fbbhjkb/gtwefprrq1j78xkizywjdbwrd8dzubhl5mxwsqpqzzinsifzzx2kmkukgh+9tkapr9eiukhtiq8buz4kcpdyjgdpoiwagz2ul+v+tk5nonmh5ngagxxtyujqjyhjshkn6xdfqxmgcusxtxwkdoa3i2ajlelzzlf9c67jhuq3v+jsnwizd2im6ehacz3abv9aecvfwcm/w4jw4t86r8zzprtjtmr34jef9c3udm9i=</latexit> <latexit sha1_base64="fdhnc2malbsxmmvjzzht9gr7lvy=">aaacd3icbzdlssnafiynxmu9rv26gsxk3zrebfw3oiugr1au8rjdnionunczeqoiw/gxldx4irt27d+tzo2oda+spax3/omznz+zfnsjvol7wvlk6tl7akg9ube/s2nv7lrulktamixgkoz4oyllim5pptjuxpcb8ttv++cqvt++pvcwk7/qkpjbw5afjia2vt8+8qijjpvikjobxzfzd3tdeakyqj4t9+2ku3omwovgflbbhrp9+9mbrcqrnnseg1jd14l1l81vjpxmzs9rnayyhihtggxbunvlp/tk+ng4axxepxq46n7eyifodre+kztgb6p+vpu/lfrjjq47kusjbnnqzj7keg41hhow8edjinrfgiaigtmr5imwasktyr5co78yovqoqu5ts29dsr18ykoejper6ikxhsb6uganvatefsantaleruerwfrzxqfts5zxcwb+ipr4xuyqzxg</latexit> <latexit sha1_base64="ehhzjncyzpsom7wi2yxooqx14gi=">aaacd3icbzdlsgmxfiyz9vbrrerstbaodvnmpkdlghvdvbax6azltjppq5ozickizzg3cooruhghifu37nwbm21bbfh8pgfc5kc3485u9q2v6zcyura+kzxs7s1vbo7v94/aksokys2smqj2fvbuc5c2tjmc9qnjqxhc9rxx1d5vxnppwjrekcnmfuedemwmalawp3yqrtiikkbg9qmol7jftgdghcqvfvzqv+u2dv7krwmzhwqak5mv/zpdikscbpqwkgpnmph2kvzmwmnwclnfi2bjgfiewzdefr56xsfdj8yz4cdsjotajx1f+kijsacn9ctajtvjlzf9qvuqhl17kwjjrncszh4keyx3hpbw8yjiszscggehm/orjcexa2ksyh+asrrwm7foay9ec23qluz/huurh6bhvkymuuandoyzqiyiebn6qa/wo/vsvvnvs9acnz85rh9kfxwdmygcsg==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> <latexit sha1_base64="zouzdqhsmxhww5fzr8opevm1do=">aaacd3icbzdlsgmxfibpek31vnxpjlgu3zqzkejoghvdvbbv6jryjs3uydizjbmhdhdn76kgxekuhxrzrfwecxmc15/chz855wk5w8swbvx3xdnanpmdm6+tfbexfpewa2srbd1ncrkwjqwsbomudpbi9yy3ah2msigmhdsirg+zusxnxphkfnzpiwrsrbxenovirv9nxq4u8xnuhqmgp6mv9gcojy52zv65v6m6nbcq+qvebkphh1co2au8+f2yppjfhgruuuo5ielm+c1usfhztzvlkf7jghusriiz7mbfpioybz+cwnlt2ri4x6fyfbqpzsb7zrorvtvwm7+v+ukjjzszjxkusmion4otauxmcndix2ugdviaagp4vavhf6hdcjycpmqvn8r/4x2fs1za95zvdqoj9oaemzcfuycbwfqgbnoqgso3mi9pmktc+c8om/oy7h1ypnmbmapoa+fnjodeq==</latexit> Sensi.vi.es (Local) Sensi.vity: par.al deriva.ve of some quan.ty of interest with respect to a parameter In our sefng, we will want to calculate sensi.vi.es of the state variables with respect to e.g. if y(t) = I(t), we would need to calculate (t) (t) How do we calculate these? If we had a formula for I(t), this would be easy... but we don t Instead we have to use the sensi.vity equa.ons

18 <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="q9i+jduzdzx6ehac4s/n4mixwq=">aaacgnicjvhlsgmxfm2mvwt9jbpeywfuagzwtcfqsgnywr2az1smplmg5p5knwryzaf4m+582sfac1fffc4hduusfjuv4iualbfjfmtcl6xmzxq7s9s7u3bxctlscssqanbax7hhemcej1gqognusyujocdb2rvetfvuzscxj6angceufzbdxgfmcmupbr24gcc38ppmhn2e3iri4efgl/8yudbmqps/vlkrbvjjwhwd8jv/w+jl1rbjaefv4ezb2urbfenp9mkyhi4akoltxsrpozrnfklheclpfekjhzmc6gkykzkqxtspmcuuzpg5iqu8eemoutmqkvgocelozehiq5d6e/k3xtsg46wu8sljgez1dfkqcq4wn+8a+l4ycggtaqot6rzgoiy4h9nzkogrn+curohvzdeyq81gr12vzoirogj2gu+sga1rhd6ibmoiid6nivils2cemy55nzoaxnzmcpo8/yt8i7hnq==</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="jp+t/fu8tyxv/tbdfqyyndzxru=">aaacdxicbvdltgixfl2dl8qx6tjni5rabmcmie4kcam7tosracf3soggdmbsdkzihb9w46+4caexbt278y/8bgearavputnhtve48bck6nbx9aqaxlldw19hpmy3nreye7u1ftfqgoq1jf+krhomace6xqubgsesighws7g4ve79+x5tmvndrrgfrs+x7vmcpmljqzi/yustaztgkcj++bm+igljgt5u8hsjm7ae9afokzi7mll5igsl+tlo+dsxzdbwoddoxa9ooktfushgmfwowib1inzvj6qfkuh1nthmt41jpkp6v4umzmlf/dqotr5jn66uaaz63kve/7xmahrn7yh7qwiyr6cp9ujbje+saeixkangmueqelxxwkdoejq4gctejz5lrdj7bto2exnppqrl6zpqboo4bdy4mazloekklafcvfwcm/wyj1yt9ar9tyttvmznn34a+v9g4vnm+=</latexit> sha1_base64="xgz29cpusm9ge8gpreseaq48rvu=">aaacdxicbvdlssnafj3uv62vqes3g1vonzwrgi4lbnrxwt6gcevmommhzirhzikub9w46+4cagiw/fu/butnqc2hhg4c869d+yel+jmacv6mgorq2vrg8xntb2zu6eux/qvmescw2rkiey64ginawpznmtbtjcsljtoonrzk/c+lymfwpycrdquma+yzajqv+uzjjxeikjobxzfts5+lmwqhyfqt6gqpb5atmjudxiz2tsoor7nvfjqdkmscbppwukpnw5f2k2w4xracmjfiybjgnjesgmqvlnjbjsppk2vafzdmz5a45n6uymbodreegmlad1si14m/uf1yu1fugkloljtgmwf8moodyizapcasuon6qeigtpxzezgqsiwczeozflzdj+7xmwzx7tl5u1pm4iugihamkstefaqbr1eqtrnadekiv6nv4nj6nn+n9xlow8p5d9afgxzfqo5q+</latexit> Sensi.vity Equa.ons Sensi.vi.es: par.al derivs of state variables with respect to params, e.g. )(t) For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters We can find the vector of sensi.vi.es of the states with respect to parameter θ i by differen.a.ng both sides of (1) with respect to the parameter, applying the chain rule, and switching the order of d/ and d/dθ i on the lem side: i with ini.al i (t) <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit i () = m <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="whrjnp69ctg2omuzoypfhupkic=">aaacghicbzdlssnafiynxmu9rv26gsxc3drecrorcm5cvraxaeqytcftmmfmroxhdygg1/fjqtf3hbn2zhpa2rrdwmf/zmhm+f3ysevwnaxsbk6tr6xwdoqb+/s7u2bb4fswsshanrcs7hlfm8jc1ging3vgyenicdbzxtv7vpdcpebtewyrm/yamq+5zskbbrnnu+jlq1imjbe4efsx+2iera+lyrgqd4wtsuwmquwbfqlkz4wwwc6igqk3xndqdicybc4ekoltptmlop/kkklhwdhlfyklhzmh6gkmsmnvpz4dl+fq7a+xhur8q8mz9pzgsqklj4onogmbildzy879alwh/qp/yme6ahxs+ye8ehgjnkeebl4ycmgggvhl9vxhrccfosuydsfephkz2hc126rzd/vko17euulh6arvkyuuqpdoizqiyqeat6q+/gs/fqfbif89yvo5g5qn9ktl8bs3qfq==</latexit> (2) : sensi9vity equa9ons (provided θ i is a pure parameter, not an ini.al condi.on- - - see later)

19 <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="q9i+jduzdzx6ehac4s/n4mixwq=">aaacgnicjvhlsgmxfm2mvwt9jbpeywfuagzwtcfqsgnywr2az1smplmg5p5knwryzaf4m+582sfac1fffc4hduusfjuv4iualbfjfmtcl6xmzxq7s9s7u3bxctlscssqanbax7hhemcej1gqognusyujocdb2rvetfvuzscxj6angceufzbdxgfmcmupbr24gcc38ppmhn2e3iri4efgl/8yudbmqps/vlkrbvjjwhwd8jv/w+jl1rbjaefv4ezb2urbfenp9mkyhi4akoltxsrpozrnfklheclpfekjhzmc6gkykzkqxtspmcuuzpg5iqu8eemoutmqkvgocelozehiq5d6e/k3xtsg46wu8sljgez1dfkqcq4wn+8a+l4ycggtaqot6rzgoiy4h9nzkogrn+curohvzdeyq81gr12vzoirogj2gu+sga1rhd6ibmoiid6nivils2cemy55nzoaxnzmcpo8/yt8i7hnq==</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="jp+t/fu8tyxv/tbdfqyyndzxru=">aaacdxicbvdltgixfl2dl8qx6tjni5rabmcmie4kcam7tosracf3soggdmbsdkzihb9w46+4caexbt278y/8bgearavputnhtve48bck6nbx9aqaxlldw19hpmy3nreye7u1ftfqgoq1jf+krhomace6xqubgsesighws7g4ve79+x5tmvndrrgfrs+x7vmcpmljqzi/yustaztgkcj++bm+igljgt5u8hsjm7ae9afokzi7mll5igsl+tlo+dsxzdbwoddoxa9ooktfushgmfwowib1inzvj6qfkuh1nthmt41jpkp6v4umzmlf/dqotr5jn66uaaz63kve/7xmahrn7yh7qwiyr6cp9ujbje+saeixkangmueqelxxwkdoejq4gctejz5lrdj7bto2exnppqrl6zpqboo4bdy4mazloekklafcvfwcm/wyj1yt9ar9tyttvmznn34a+v9g4vnm+=</latexit> sha1_base64="xgz29cpusm9ge8gpreseaq48rvu=">aaacdxicbvdlssnafj3uv62vqes3g1vonzwrgi4lbnrxwt6gcevmommhzirhzikub9w46+4cagiw/fu/butnqc2hhg4c869d+yel+jmacv6mgorq2vrg8xntb2zu6eux/qvmescw2rkiey64ginawpznmtbtjcsljtoonrzk/c+lymfwpycrdquma+yzajqv+uzjjxeikjobxzfts5+lmwqhyfqt6gqpb5atmjudxiz2tsoor7nvfjqdkmscbppwukpnw5f2k2w4xracmjfiybjgnjesgmqvlnjbjsppk2vafzdmz5a45n6uymbodreegmlad1si14m/uf1yu1fugkloljtgmwf8moodyizapcasuon6qeigtpxzezgqsiwczeozflzdj+7xmwzx7tl5u1pm4iugihamkstefaqbr1eqtrnadekiv6nv4nj6nn+n9xlow8p5d9afgxzfqo5q+</latexit> Sensi.vity Equa.ons Sensi.vi.es: par.al derivs of state variables with respect to params, e.g. )(t) For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters We can find the vector of sensi.vi.es of the states with respect to parameter θ i by differen.a.ng both sides of (1) with respect to the parameter, applying the chain rule, and switching the order of d/ and d/dθ i on the lem side: i with ini.al i (t) <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit i () = m <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="whrjnp69ctg2omuzoypfhupkic=">aaacghicbzdlssnafiynxmu9rv26gsxc3drecrorcm5cvraxaeqytcftmmfmroxhdygg1/fjqtf3hbn2zhpa2rrdwmf/zmhm+f3ysevwnaxsbk6tr6xwdoqb+/s7u2bb4fswsshanrcs7hlfm8jc1ging3vgyenicdbzxtv7vpdcpebtewyrm/yamq+5zskbbrnnu+jlq1imjbe4efsx+2iera+lyrgqd4wtsuwmquwbfqlkz4wwwc6igqk3xndqdicybc4ekoltptmlop/kkklhwdhlfyklhzmh6gkmsmnvpz4dl+fq7a+xhur8q8mz9pzgsqklj4onogmbildzy879alwh/qp/yme6ahxs+ye8ehgjnkeebl4ycmgggvhl9vxhrccfosuydsfephkz2hc126rzd/vko17euulh6arvkyuuqpdoizqiyqeat6q+/gs/fqfbif89yvo5g5qn9ktl8bs3qfq==</latexit> (2) : sensi9vity equa9ons Why do we need the chain rule? Because solu.on x(t) depends on (provided θ i is a pure parameter, not an ini.al condi.on- - - parameter see later) θ i

20 <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="q9i+jduzdzx6ehac4s/n4mixwq=">aaacgnicjvhlsgmxfm2mvwt9jbpeywfuagzwtcfqsgnywr2az1smplmg5p5knwryzaf4m+582sfac1fffc4hduusfjuv4iualbfjfmtcl6xmzxq7s9s7u3bxctlscssqanbax7hhemcej1gqognusyujocdb2rvetfvuzscxj6angceufzbdxgfmcmupbr24gcc38ppmhn2e3iri4efgl/8yudbmqps/vlkrbvjjwhwd8jv/w+jl1rbjaefv4ezb2urbfenp9mkyhi4akoltxsrpozrnfklheclpfekjhzmc6gkykzkqxtspmcuuzpg5iqu8eemoutmqkvgocelozehiq5d6e/k3xtsg46wu8sljgez1dfkqcq4wn+8a+l4ycggtaqot6rzgoiy4h9nzkogrn+curohvzdeyq81gr12vzoirogj2gu+sga1rhd6ibmoiid6nivils2cemy55nzoaxnzmcpo8/yt8i7hnq==</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="jp+t/fu8tyxv/tbdfqyyndzxru=">aaacdxicbvdltgixfl2dl8qx6tjni5rabmcmie4kcam7tosracf3soggdmbsdkzihb9w46+4caexbt278y/8bgearavputnhtve48bck6nbx9aqaxlldw19hpmy3nreye7u1ftfqgoq1jf+krhomace6xqubgsesighws7g4ve79+x5tmvndrrgfrs+x7vmcpmljqzi/yustaztgkcj++bm+igljgt5u8hsjm7ae9afokzi7mll5igsl+tlo+dsxzdbwoddoxa9ooktfushgmfwowib1inzvj6qfkuh1nthmt41jpkp6v4umzmlf/dqotr5jn66uaaz63kve/7xmahrn7yh7qwiyr6cp9ujbje+saeixkangmueqelxxwkdoejq4gctejz5lrdj7bto2exnppqrl6zpqboo4bdy4mazloekklafcvfwcm/wyj1yt9ar9tyttvmznn34a+v9g4vnm+=</latexit> sha1_base64="xgz29cpusm9ge8gpreseaq48rvu=">aaacdxicbvdlssnafj3uv62vqes3g1vonzwrgi4lbnrxwt6gcevmommhzirhzikub9w46+4cagiw/fu/butnqc2hhg4c869d+yel+jmacv6mgorq2vrg8xntb2zu6eux/qvmescw2rkiey64ginawpznmtbtjcsljtoonrzk/c+lymfwpycrdquma+yzajqv+uzjjxeikjobxzfts5+lmwqhyfqt6gqpb5atmjudxiz2tsoor7nvfjqdkmscbppwukpnw5f2k2w4xracmjfiybjgnjesgmqvlnjbjsppk2vafzdmz5a45n6uymbodreegmlad1si14m/uf1yu1fugkloljtgmwf8moodyizapcasuon6qeigtpxzezgqsiwczeozflzdj+7xmwzx7tl5u1pm4iugihamkstefaqbr1eqtrnadekiv6nv4nj6nn+n9xlow8p5d9afgxzfqo5q+</latexit> Sensi.vity Equa.ons Sensi.vi.es: par.al derivs of state variables with respect to params, e.g. )(t) For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters We can find the vector of sensi.vi.es of the states with respect to parameter θ i by differen.a.ng both sides of (1) with respect to the parameter, applying the chain rule, and switching the order of d/ and d/dθ i on the lem side: i with ini.al i (t) <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit i () = m <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="whrjnp69ctg2omuzoypfhupkic=">aaacghicbzdlssnafiynxmu9rv26gsxc3drecrorcm5cvraxaeqytcftmmfmroxhdygg1/fjqtf3hbn2zhpa2rrdwmf/zmhm+f3ysevwnaxsbk6tr6xwdoqb+/s7u2bb4fswsshanrcs7hlfm8jc1ging3vgyenicdbzxtv7vpdcpebtewyrm/yamq+5zskbbrnnu+jlq1imjbe4efsx+2iera+lyrgqd4wtsuwmquwbfqlkz4wwwc6igqk3xndqdicybc4ekoltptmlop/kkklhwdhlfyklhzmh6gkmsmnvpz4dl+fq7a+xhur8q8mz9pzgsqklj4onogmbildzy879alwh/qp/yme6ahxs+ye8ehgjnkeebl4ycmgggvhl9vxhrccfosuydsfephkz2hc126rzd/vko17euulh6arvkyuuqpdoizqiyqeat6q+/gs/fqfbif89yvo5g5qn9ktl8bs3qfq==</latexit> (2) : sensi9vity equa9ons Why are these zero? Ini.al condi.ons don t depend on parameter (provided θ i is a pure parameter, not an ini.al condi.on- - - value see later)

21 <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="q9i+jduzdzx6ehac4s/n4mixwq=">aaacgnicjvhlsgmxfm2mvwt9jbpeywfuagzwtcfqsgnywr2az1smplmg5p5knwryzaf4m+582sfac1fffc4hduusfjuv4iualbfjfmtcl6xmzxq7s9s7u3bxctlscssqanbax7hhemcej1gqognusyujocdb2rvetfvuzscxj6angceufzbdxgfmcmupbr24gcc38ppmhn2e3iri4efgl/8yudbmqps/vlkrbvjjwhwd8jv/w+jl1rbjaefv4ezb2urbfenp9mkyhi4akoltxsrpozrnfklheclpfekjhzmc6gkykzkqxtspmcuuzpg5iqu8eemoutmqkvgocelozehiq5d6e/k3xtsg46wu8sljgez1dfkqcq4wn+8a+l4ycggtaqot6rzgoiy4h9nzkogrn+curohvzdeyq81gr12vzoirogj2gu+sga1rhd6ibmoiid6nivils2cemy55nzoaxnzmcpo8/yt8i7hnq==</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="bqs2f9rcai8w9gtee9nkiye59e=">aaacdxicbzdlsgmxfibpek31vnxpjlgfv2vgcrqz4mzlbxubzjbkkwbmrmqnbhkbdw46u4cagiw/fufasfwfqcauspgy//njpk/eeqhubb/rswlldw19ylg8xnre2d3dleflmnmwk8wrkzqhzanzci5gukhk7vzxggestyha1rrfuuniiiw9xmhivor1yhijrnjzfonzdrvnuplshojkeox92sc+r+mlkl8p2xz6iliizg/llfxu9sfbjdhwcrjzjjq3xhsfl18fc+tffrm81tygaxzsgyxpx7ewtbubkxdhdeibknbjjxp9kdni62eumm6iyl/p18bmf7vohugfl4s4zzdhbppqmemccrlhq7pccyzyaiayjcxfcettew+aaismbgd+5uvonlucu+lcvmu16jqnkmahhmepohaonbigojsawt8wjo8wa/wk/vqvu1bl6zzzah8kfx+dq7any4=</latexit> <latexit sha1_base64="uo94mfus6d+gi/nqunshwv7zr8=">aaacdxicbzdlssnafiynxmu9rv26gaycq5jiqzcfny4r2asjuymk3bozbjmtoqs8gjufbu3lhrx696db+okdaitpwx8/oecmtl/kaiuwxg+rjxvtfwnzcpwdxtnd2/fpjjs6dhvllvplglvc4hmgkvwbg6c9rlfsbqi1gm1w9e8+u5rg8g2ncbhezsr5yssbyvn3qhyrqzeuiak4edvmf9mdmgpg89+2au3dmwsvgllbdpvq+/ekny5pgtaivrou+6yqwyip7qwb51us1swidkbhrg5qkynqqzbbj8zlxhjimltks8mz9pzgrsotpfjjoimbyl9yk879ap4xwapbxmatajj/fkycq4ylapcqkzbtaqqrj5k6zjyuibe2dvhoaurrwmnyu669td2at2sjjqkbjdilokysuurpdobzqi4oebn6qa/wo/vsvvnv89yvq5w5qn9kfxwdc66cxw==</latexit> <latexit sha1_base64="bqs2f9rcai8w9gtee9nkiye59e=">aaacdxicbzdlsgmxfibpek31vnxpjlgfv2vgcrqz4mzlbxubzjbkkwbmrmqnbhkbdw46u4cagiw/fufasfwfqcauspgy//njpk/eeqhubb/rswlldw19ylg8xnre2d3dleflmnmwk8wrkzqhzanzci5gukhk7vzxggestyha1rrfuuniiiw9xmhivor1yhijrnjzfonzdrvnuplshojkeox92sc+r+mlkl8p2xz6iliizg/llfxu9sfbjdhwcrjzjjq3xhsfl18fc+tffrm81tygaxzsgyxpx7ewtbubkxdhdeibknbjjxp9kdni62eumm6iyl/p18bmf7vohugfl4s4zzdhbppqmemccrlhq7pccyzyaiayjcxfcettew+aaismbgd+5uvonlucu+lcvmu16jqnkmahhmepohaonbigojsawt8wjo8wa/wk/vqvu1bl6zzzah8kfx+dq7any4=</latexit> <latexit sha1_base64="bqs2f9rcai8w9gtee9nkiye59e=">aaacdxicbzdlsgmxfibpek31vnxpjlgfv2vgcrqz4mzlbxubzjbkkwbmrmqnbhkbdw46u4cagiw/fufasfwfqcauspgy//njpk/eeqhubb/rswlldw19ylg8xnre2d3dleflmnmwk8wrkzqhzanzci5gukhk7vzxggestyha1rrfuuniiiw9xmhivor1yhijrnjzfonzdrvnuplshojkeox92sc+r+mlkl8p2xz6iliizg/llfxu9sfbjdhwcrjzjjq3xhsfl18fc+tffrm81tygaxzsgyxpx7ewtbubkxdhdeibknbjjxp9kdni62eumm6iyl/p18bmf7vohugfl4s4zzdhbppqmemccrlhq7pccyzyaiayjcxfcettew+aaismbgd+5uvonlucu+lcvmu16jqnkmahhmepohaonbigojsawt8wjo8wa/wk/vqvu1bl6zzzah8kfx+dq7any4=</latexit> <latexit sha1_base64="bqs2f9rcai8w9gtee9nkiye59e=">aaacdxicbzdlsgmxfibpek31vnxpjlgfv2vgcrqz4mzlbxubzjbkkwbmrmqnbhkbdw46u4cagiw/fufasfwfqcauspgy//njpk/eeqhubb/rswlldw19ylg8xnre2d3dleflmnmwk8wrkzqhzanzci5gukhk7vzxggestyha1rrfuuniiiw9xmhivor1yhijrnjzfonzdrvnuplshojkeox92sc+r+mlkl8p2xz6iliizg/llfxu9sfbjdhwcrjzjjq3xhsfl18fc+tffrm81tygaxzsgyxpx7ewtbubkxdhdeibknbjjxp9kdni62eumm6iyl/p18bmf7vohugfl4s4zzdhbppqmemccrlhq7pccyzyaiayjcxfcettew+aaismbgd+5uvonlucu+lcvmu16jqnkmahhmepohaonbigojsawt8wjo8wa/wk/vqvu1bl6zzzah8kfx+dq7any4=</latexit> <latexit sha1_base64="jp+t/fu8tyxv/tbdfqyyndzxru=">aaacdxicbvdltgixfl2dl8qx6tjni5rabmcmie4kcam7tosracf3soggdmbsdkzihb9w46+4caexbt278y/8bgearavputnhtve48bck6nbx9aqaxlldw19hpmy3nreye7u1ftfqgoq1jf+krhomace6xqubgsesighws7g4ve79+x5tmvndrrgfrs+x7vmcpmljqzi/yustaztgkcj++bm+igljgt5u8hsjm7ae9afokzi7mll5igsl+tlo+dsxzdbwoddoxa9ooktfushgmfwowib1inzvj6qfkuh1nthmt41jpkp6v4umzmlf/dqotr5jn66uaaz63kve/7xmahrn7yh7qwiyr6cp9ujbje+saeixkangmueqelxxwkdoejq4gctejz5lrdj7bto2exnppqrl6zpqboo4bdy4mazloekklafcvfwcm/wyj1yt9ar9tyttvmznn34a+v9g4vnm+=</latexit> sha1_base64="xgz29cpusm9ge8gpreseaq48rvu=">aaacdxicbvdlssnafj3uv62vqes3g1vonzwrgi4lbnrxwt6gcevmommhzirhzikub9w46+4cagiw/fu/butnqc2hhg4c869d+yel+jmacv6mgorq2vrg8xntb2zu6eux/qvmescw2rkiey64ginawpznmtbtjcsljtoonrzk/c+lymfwpycrdquma+yzajqv+uzjjxeikjobxzfts5+lmwqhyfqt6gqpb5atmjudxiz2tsoor7nvfjqdkmscbppwukpnw5f2k2w4xracmjfiybjgnjesgmqvlnjbjsppk2vafzdmz5a45n6uymbodreegmlad1si14m/uf1yu1fugkloljtgmwf8moodyizapcasuon6qeigtpxzezgqsiwczeozflzdj+7xmwzx7tl5u1pm4iugihamkstefaqbr1eqtrnadekiv6nv4nj6nn+n9xlow8p5d9afgxzfqo5q+</latexit> Sensi.vity Equa.ons Sensi.vi.es: par.al derivs of state variables with respect to params, e.g. )(t) For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters We can find the vector of i (t) <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> <latexit sha1_base64="lqgurfr7xeewqju598rclf/sjp=">aaacehicbzdlssnafiyn9vbrlerszwar66ykutblwy3lcvyctqit6aqdorkwcykwkedw46u4cagiw5fufbunbrbt/whg4z/nzmz5/urwbzb1zzrwvtfwn8qbla3tnd9c/+go+juutamsyhlzyekcr6xnnaqrjdirkjfsk4/vprwu3dmkh5htzbjmbusycqdtgloyznpnuasmjkjkccjwpf5dzswyka8ntfgdhtm1apbm+flsauookitz/xbjfnqxybfuspvml4gbtq6lgecvjfusihzmh62umsmium8wyvgjdgy4iku+eecz+3sii6fsk9dxnsgbkvqstc3/avugks341gsaovo/kegfrhipehd7hkfmrea6gs679ioii6idazvnqi9ulky9a5r9tw3b5pvjunio4yoklhqizsdiga6bq1ubtr9iceat6nr6nz+pnej+3loxi5hd9kfhxdvn1nvq=</latexit> of the states with respect to parameter θ i by differen.a.ng both sides of (1) with respect to the parameter, applying the chain rule, and switching the order of d/ and d/dθ i on the lem side: The is the Jacobian matrix d (2) : sensi9vity - differen.ate equa9ons RHS of ODE w.r.t. state i with ini.al is deriva.ve of RHS w.r.t. parameter () i <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> sha1_base64="whrjnp69ctg2omuzoypfhupkic=">aaacghicbzdlssnafiynxmu9rv26gsxc3drecrorcm5cvraxaeqytcftmmfmroxhdygg1/fjqtf3hbn2zhpa2rrdwmf/zmhm+f3ysevwnaxsbk6tr6xwdoqb+/s7u2bb4fswsshanrcs7hlfm8jc1ging3vgyenicdbzxtv7vpdcpebtewyrm/yamq+5zskbbrnnu+jlq1imjbe4efsx+2iera+lyrgqd4wtsuwmquwbfqlkz4wwwc6igqk3xndqdicybc4ekoltptmlop/kkklhwdhlfyklhzmh6gkmsmnvpz4dl+fq7a+xhur8q8mz9pzgsqklj4onogmbildzy879alwh/qp/yme6ahxs+ye8ehgjnkeebl4ycmgggvhl9vxhrccfosuydsfephkz2hc126rzd/vko17euulh6arvkyuuqpdoizqiyqeat6q+/gs/fqfbif89yvo5g5qn9ktl8bs3qfq==</latexit> i This is a generaliza.on of linear stability (provided θ i is a pure parameter, not an ini.al analysis condi.on- - - see later)

22 <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="whrjnp69ctg2omuzoypfhupkic=">aaacghicbzdlssnafiynxmu9rv26gsxc3drecrorcm5cvraxaeqytcftmmfmroxhdygg1/fjqtf3hbn2zhpa2rrdwmf/zmhm+f3ysevwnaxsbk6tr6xwdoqb+/s7u2bb4fswsshanrcs7hlfm8jc1ging3vgyenicdbzxtv7vpdcpebtewyrm/yamq+5zskbbrnnu+jlq1imjbe4efsx+2iera+lyrgqd4wtsuwmquwbfqlkz4wwwc6igqk3xndqdicybc4ekoltptmlop/kkklhwdhlfyklhzmh6gkmsmnvpz4dl+fq7a+xhur8q8mz9pzgsqklj4onogmbildzy879alwh/qp/yme6ahxs+ye8ehgjnkeebl4ycmgggvhl9vxhrccfosuydsfephkz2hc126rzd/vko17euulh6arvkyuuqpdoizqiyqeat6q+/gs/fqfbif89yvo5g5qn9ktl8bs3qfq==</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="o8ckqtnsexod8rscmcgexs+opxm=">aaacghicbzdjsgnbeizrxgpcoh69nazbl3fgbl2ighepecwcmtddhqsjjl3tvigoyxvpgqxjwo4ju338jhsloas/yh4eovkqrr9xmpnnr2hzu3v7c4tfxyka6urw9slra26zpofem1fstynx2qurqrr6fayzuj4jtjw/4/atrvxhhlrzxdiudhldd2o1eibhfy3mlizdqlgvuqhukksl9/su9jhst+qh9ie5j7axhblxktsveywyc84uyhefmfbvkw3dtszskefijnw65dgjtrprciz5xnrtzrpk+rtlwwyjgnldzsah5wtfobsxmq8cmny/tmrvdrqeibzpbit/+tjcz/aqug7n2jqikrr6xyaiglqrjmkqjditidoxaagvkml8s1qmmktrzfkizt+tz6f+xhhsinnzur48maqbbdifptgab7heq6hcjvg8abp8akv1qp1bl1z75pwows6swo/za2/aoz9oqa=</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="q9i+jduzdzx6ehac4s/n4mixwq=">aaacgnicjvhlsgmxfm2mvwt9jbpeywfuagzwtcfqsgnywr2az1smplmg5p5knwryzaf4m+582sfac1fffc4hduusfjuv4iualbfjfmtcl6xmzxq7s9s7u3bxctlscssqanbax7hhemcej1gqognusyujocdb2rvetfvuzscxj6angceufzbdxgfmcmupbr24gcc38ppmhn2e3iri4efgl/8yudbmqps/vlkrbvjjwhwd8jv/w+jl1rbjaefv4ezb2urbfenp9mkyhi4akoltxsrpozrnfklheclpfekjhzmc6gkykzkqxtspmcuuzpg5iqu8eemoutmqkvgocelozehiq5d6e/k3xtsg46wu8sljgez1dfkqcq4wn+8a+l4ycggtaqot6rzgoiy4h9nzkogrn+curohvzdeyq81gr12vzoirogj2gu+sga1rhd6ibmoiid6nivils2cemy55nzoaxnzmcpo8/yt8i7hnq==</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="fqh6uyux6yh3g7+i79cws8gpoqu=">aaacgnicjvhlsgmxfm2mvwt9vv26czacknqzlehcsedgzqx7ge4pmuymdc8so6izzgp8bfc+rf+gayptlyixggczj33jdnxjqvxyflvhrmsw11bz28unre2d3ale/tnfswssganrctbllfm8ja1ging7vgyerictdzh3bjfemjs8sh8hfhmughph9znlicmesuxx5eepl6wepbnsrmtczwi/jx9ywcgdeipzzclij+bh/ipaz7ff3h8yxrfklwxjowxgtdp3nkl6r/jqebfnahycfuspjm3fe3hvlswroakiswedkmfdtqmscbun51emogyzjzsr1kfepcenz9isadukhc1miawuiu9mflbr5oaf9vnergnwei6vchpbiyij/ebps4zbthsgfdj9vsxhradd+itfxqi9ukxlhzvgjbffuhwqpvp2mgpdper+gy2egs1da9qqmgoujdkbsv48zmmsembv5mpayxmzlap8q8/gsnqmhk</latexit> <latexit sha1_base64="jp+t/fu8tyxv/tbdfqyyndzxru=">aaacdxicbvdltgixfl2dl8qx6tjni5rabmcmie4kcam7tosracf3soggdmbsdkzihb9w46+4caexbt278y/8bgearavputnhtve48bck6nbx9aqaxlldw19hpmy3nreye7u1ftfqgoq1jf+krhomace6xqubgsesighws7g4ve79+x5tmvndrrgfrs+x7vmcpmljqzi/yustaztgkcj++bm+igljgt5u8hsjm7ae9afokzi7mll5igsl+tlo+dsxzdbwoddoxa9ooktfushgmfwowib1inzvj6qfkuh1nthmt41jpkp6v4umzmlf/dqotr5jn66uaaz63kve/7xmahrn7yh7qwiyr6cp9ujbje+saeixkangmueqelxxwkdoejq4gctejz5lrdj7bto2exnppqrl6zpqboo4bdy4mazloekklafcvfwcm/wyj1yt9ar9tyttvmznn34a+v9g4vnm+=</latexit> sha1_base64="xgz29cpusm9ge8gpreseaq48rvu=">aaacdxicbvdlssnafj3uv62vqes3g1vonzwrgi4lbnrxwt6gcevmommhzirhzikub9w46+4cagiw/fu/butnqc2hhg4c869d+yel+jmacv6mgorq2vrg8xntb2zu6eux/qvmescw2rkiey64ginawpznmtbtjcsljtoonrzk/c+lymfwpycrdquma+yzajqv+uzjjxeikjobxzfts5+lmwqhyfqt6gqpb5atmjudxiz2tsoor7nvfjqdkmscbppwukpnw5f2k2w4xracmjfiybjgnjesgmqvlnjbjsppk2vafzdmz5a45n6uymbodreegmlad1si14m/uf1yu1fugkloljtgmwf8moodyizapcasuon6qeigtpxzezgqsiwczeozflzdj+7xmwzx7tl5u1pm4iugihamkstefaqbr1eqtrnadekiv6nv4nj6nn+n9xlow8p5d9afgxzfqo5q+</latexit> Sensi.vity Equa.ons Sensi.vi.es: par.al derivs of state variables with respect to params, e.g. )(t) For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters The sensi.vity equa.ons for <latexit sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> sha1_base64="9rctmhmhmax5gl5oa45z/42rsg=">aaacehicbzdlsgmxfibp1futt6pln8ei1k2zkylullhxwcfeofnkjs2ozklyrmxdhen76kgxekuhxpzrfweuwvilb+epj4zzljzu/fumi7u8rs7s8srqwxc9tbg5t7+r39+o6shtjnrbjsdu9qrkuia+hqmmbsei8crveiplcb1xy5uwuxidw5i3a9olhs8yrwn18seuryhl3zgqffssu9epu9jnsdtivmqtskx7ji9evkezwafiy+yqnrjf7jdicubd5fjqnxlswnsp+ormesjnjtohlm2od3emhjsgot2ollori6myv+pmwjkuzc3xmpdbqebp7pdcj29xxtbp5xayxon7dtecyj8pbnh/itstai43rivyjoua4nukae+sthfwosqpnhzotgzk+8cpxtkmoxnotyovkepgfzoibdkiidz1cbk6hcdrjcwym8w4v1yd1zr9bbtdvjzwb24y+s92/+j6d</latexit> sha1_base64="lqgurfr7xeewqju598rclf/sjp=">aaacehicbzdlssnafiyn9vbrlerszwar66ykutblwy3lcvyctqit6aqdorkwcykwkedw46u4cagiw5fufbunbrbt/whg4z/nzmz5/urwbzb1zzrwvtfwn8qbla3tnd9c/+go+juutamsyhlzyekcr6xnnaqrjdirkjfsk4/vprwu3dmkh5htzbjmbusycqdtgloyznpnuasmjkjkccjwpf5dzswyka8ntfgdhtm1apbm+flsauookitz/xbjfnqxybfuspvml4gbtq6lgecvjfusihzmh62umsmium8wyvgjdgy4iku+eecz+3sii6fsk9dxnsgbkvqstc3/avugks341gsaovo/kegfrhipehd7hkfmrea6gs679ioii6idazvnqi9ulky9a5r9tw3b5pvjunio4yoklhqizsdiga6bq1ubtr9iceat6nr6nz+pnej+3loxi5hd9kfhxdvn1nvq=</latexit> i with ini.al condi.ons () = i are i 1. Sensi.vity eqns for different states with respect to a given parameter are coupled 2. Sensi.vity eqns across different parameters are not coupled so can arrange sensi.vi.es with respect to different parameters into a matrix

23 Sensi.vity Equa.ons For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters the m by p matrix of sensi.vi.es, with ini.al () = m p, sa.sfies the ODE system Equa.ons (2) are the sensi.vity equa.ons for the system (2) Sensi9vity matrix: Sensi.vi.es of different states with respect to a given parameter are arranged in a single column, different columns depict sensi.vi.es with respect to different parameters

24 <latexit sha1_base64="2wz86xapxcdloap+oek77ulhtw=">aaacbnicbzdlssnafizp6q3ww9slcinfcfuskejoghuxfewfmlamk7dhjhzikwkjubx8wncxc+gzufasfwulauft/gpj4zzkzc34v5kwqy/oskvlk6tr5fxkxubw9o65u9ewusiibzgir6lryuk5c2llmcvpnxyubx6nhw98mdc7t1rifou3ahjtn8ddkpmmykwtvnno+akt1imxuaxz5gc/fjf1zapvswqhrbbnul34gklnvvnhdcksbdruhgmpe7yvkzfnlysczhunkttgziyhtkcxxagvblqskafj7qyqhwl9qouk9/deigmpj4gnowosrnk+lpv/1xqj8s/dlivxomhipg/5cucqqnkmamaejyppngaimp4riiosc1e6uyoowz5ferhapzxbqtnx9wqjpkdynaar3acnpxba66gcsgca+p8awvxopxzlwab9pwkjgb2yc/mt6/aymlmoy=</latexit> <latexit sha1_base64="hkhfpsd37t7mbqsjksbpdanqyo=">aaacbnicbzdlssnafizp6q3ww9slcinfcfuskeiy4mzlbxubjptjdnionuzczeqsoss3voobf4q49rnc+tzo2oda+spax3/omznzbwlnsjvol1vawv1b3yhvvra2d3b37p2dtoptswilxdyw3qarypmglcp91euhwfnhac8vve79xrqvgsbvukox6eh4kfjgbtrl597iusk8xlsnqmcxrof/h+2rerts2zcs2dwavcjx79qc3iekauaejxr1xcfrfpzfsdidvrxuqstmr7snkgbi6r8blbgfjaz4dcwjojnjq5vycyhck1iqltgwe9uou13pyv1kt1eolntcsppolmhwptjnsm8kzqgelknj8yweqy81dertjkok1yfrocu7jymrtpa65tc2/q1ua9ikmmr3acz+dcbttggprqagip8aqv8go9ws/wm/u+by1zxcwh/jh18q3n+zlx</latexit> <latexit sha1_base64="2wz86xapxcdloap+oek77ulhtw=">aaacbnicbzdlssnafizp6q3ww9slcinfcfuskejoghuxfewfmlamk7dhjhzikwkjubx8wncxc+gzufasfwulauft/gpj4zzkzc34v5kwqy/oskvlk6tr5fxkxubw9o65u9ewusiibzgir6lryuk5c2llmcvpnxyubx6nhw98mdc7t1rifou3ahjtn8ddkpmmykwtvnno+akt1imxuaxz5gc/fjf1zapvswqhrbbnul34gklnvvnhdcksbdruhgmpe7yvkzfnlysczhunkttgziyhtkcxxagvblqskafj7qyqhwl9qouk9/deigmpj4gnowosrnk+lpv/1xqj8s/dlivxomhipg/5cucqqnkmamaejyppngaimp4riiosc1e6uyoowz5ferhapzxbqtnx9wqjpkdynaar3acnpxba66gcsgca+p8awvxopxzlwab9pwkjgb2yc/mt6/aymlmoy=</latexit> <latexit sha1_base64="2wz86xapxcdloap+oek77ulhtw=">aaacbnicbzdlssnafizp6q3ww9slcinfcfuskejoghuxfewfmlamk7dhjhzikwkjubx8wncxc+gzufasfwulauft/gpj4zzkzc34v5kwqy/oskvlk6tr5fxkxubw9o65u9ewusiibzgir6lryuk5c2llmcvpnxyubx6nhw98mdc7t1rifou3ahjtn8ddkpmmykwtvnno+akt1imxuaxz5gc/fjf1zapvswqhrbbnul34gklnvvnhdcksbdruhgmpe7yvkzfnlysczhunkttgziyhtkcxxagvblqskafj7qyqhwl9qouk9/deigmpj4gnowosrnk+lpv/1xqj8s/dlivxomhipg/5cucqqnkmamaejyppngaimp4riiosc1e6uyoowz5ferhapzxbqtnx9wqjpkdynaar3acnpxba66gcsgca+p8awvxopxzlwab9pwkjgb2yc/mt6/aymlmoy=</latexit> <latexit sha1_base64="2wz86xapxcdloap+oek77ulhtw=">aaacbnicbzdlssnafizp6q3ww9slcinfcfuskejoghuxfewfmlamk7dhjhzikwkjubx8wncxc+gzufasfwulauft/gpj4zzkzc34v5kwqy/oskvlk6tr5fxkxubw9o65u9ewusiibzgir6lryuk5c2llmcvpnxyubx6nhw98mdc7t1rifou3ahjtn8ddkpmmykwtvnno+akt1imxuaxz5gc/fjf1zapvswqhrbbnul34gklnvvnhdcksbdruhgmpe7yvkzfnlysczhunkttgziyhtkcxxagvblqskafj7qyqhwl9qouk9/deigmpj4gnowosrnk+lpv/1xqj8s/dlivxomhipg/5cucqqnkmamaejyppngaimp4riiosc1e6uyoowz5ferhapzxbqtnx9wqjpkdynaar3acnpxba66gcsgca+p8awvxopxzlwab9pwkjgb2yc/mt6/aymlmoy=</latexit> Sensi.vity Equa.ons For dx = f(x, t; ) (1), where x and f are m dimensional (m state variables), and θ is a p dimensional vector of parameters the m by p matrix of sensi.vi.es, with ini.al () = m p, sa.sfies the ODE system Equa.ons (2) are the sensi.vity equa.ons for the system The Jacobian matrix depends on the state variables, so the sensi.vity equa.ons (2) [a linear system of ODEs] must be solved simultaneously with the governing equa.ons (1)

25 <latexit sha1_base64="tg2ytmfgfrmwqpiwfp2wpldfera=">aaacxhicjzhlsgmxfiyzy9verfyfn26crxbvzqsgg7hgxmufe4fokzlmpo3nxejoigwyl3txjc/gg2g6lvhbfx4ifpznlvxxy8evwnbcmhcku3v7xvk5cla9pkodn3rvlejkojqskey7rdhbq9ybdol1y8li4arwc6cpi3zvlunfo/azzjebbmqccp9taloa1ztjsjtls9yjbsxeqcjwk/zt/swiqbgb1kdxtffrbe8j8jefsrwwrd7wn9grq958oj/ao9u54eucfgivrkmbbcuwtbedqwbz2ukuiwmdkjebaaxjwnqwzc3j8kvwpoxhup8qck6ud6qkugowuloyidbrm7mf+fdukib/ox5gcfaqrpc5cccq4qxtmopszbzdqqkrm+k6ytovb/r9lbyk9+ert6f43bkthpzxrrebsdvre5+gcxseb3aawekrt1eeuzdgxutrkxodzmctmdvlqgqueu/qrzlnvmmw6qw==</latexit> <latexit sha1_base64="ptgeob/x1+zhsbvguwabd3fp8q=">aaacxhicjzfns8mwgmftonuvtqucfy/bixgarqzigy8ejzgxmadi3tls59ixkqjtiv6wxv4pmw8g5efcbwi//85b84ywck7dthwhulfypdsuvau2ofnxinz71vjxkyrofreceeqxwspwbq6cdrljsogj1vdmj8t8/41jxepobeyjg4vkevgauwjaglvkdsshmz9npurrdhmigrob3/mfdmhkgixf84etoidfbpjpbdy2gnbtxgxebaeabiqim7y+xd+macgioiionxtsbebzcjavlk+6qwijotmyyuoneqmzgmurc3j8rrufb7hujwk8ujc7mhiqnq89xrksmkrt3fl8kzdmibgfztxkumarxs8kuoehxkunsc8loydmggivxn8vynr/od+j6o2wdl+8i7bpuo3xsew412q7cjjc7rfbpbdrpdbfseoqilkfqgl6nsvixps2twzpq61dsknnpk8ylb5dturq=</latexit> <latexit sha1_base64="tg2ytmfgfrmwqpiwfp2wpldfera=">aaacxhicjzhlsgmxfiyzy9verfyfn26crxbvzqsgg7hgxmufe4fokzlmpo3nxejoigwyl3txjc/gg2g6lvhbfx4ifpznlvxxy8evwnbcmhcku3v7xvk5cla9pkodn3rvlejkojqskey7rdhbq9ybdol1y8li4arwc6cpi3zvlunfo/azzjebbmqccp9taloa1ztjsjtls9yjbsxeqcjwk/zt/swiqbgb1kdxtffrbe8j8jefsrwwrd7wn9grq958oj/ao9u54eucfgivrkmbbcuwtbedqwbz2ukuiwmdkjebaaxjwnqwzc3j8kvwpoxhup8qck6ud6qkugowuloyidbrm7mf+fdukib/ox5gcfaqrpc5cccq4qxtmopszbzdqqkrm+k6ytovb/r9lbyk9+ert6f43bkthpzxrrebsdvre5+gcxseb3aawekrt1eeuzdgxutrkxodzmctmdvlqgqueu/qrzlnvmmw6qw==</latexit> <latexit sha1_base64="tg2ytmfgfrmwqpiwfp2wpldfera=">aaacxhicjzhlsgmxfiyzy9verfyfn26crxbvzqsgg7hgxmufe4fokzlmpo3nxejoigwyl3txjc/gg2g6lvhbfx4ifpznlvxxy8evwnbcmhcku3v7xvk5cla9pkodn3rvlejkojqskey7rdhbq9ybdol1y8li4arwc6cpi3zvlunfo/azzjebbmqccp9taloa1ztjsjtls9yjbsxeqcjwk/zt/swiqbgb1kdxtffrbe8j8jefsrwwrd7wn9grq958oj/ao9u54eucfgivrkmbbcuwtbedqwbz2ukuiwmdkjebaaxjwnqwzc3j8kvwpoxhup8qck6ud6qkugowuloyidbrm7mf+fdukib/ox5gcfaqrpc5cccq4qxtmopszbzdqqkrm+k6ytovb/r9lbyk9+ert6f43bkthpzxrrebsdvre5+gcxseb3aawekrt1eeuzdgxutrkxodzmctmdvlqgqueu/qrzlnvmmw6qw==</latexit> <latexit sha1_base64="tg2ytmfgfrmwqpiwfp2wpldfera=">aaacxhicjzhlsgmxfiyzy9verfyfn26crxbvzqsgg7hgxmufe4fokzlmpo3nxejoigwyl3txjc/gg2g6lvhbfx4ifpznlvxxy8evwnbcmhcku3v7xvk5cla9pkodn3rvlejkojqskey7rdhbq9ybdol1y8li4arwc6cpi3zvlunfo/azzjebbmqccp9taloa1ztjsjtls9yjbsxeqcjwk/zt/swiqbgb1kdxtffrbe8j8jefsrwwrd7wn9grq958oj/ao9u54eucfgivrkmbbcuwtbedqwbz2ukuiwmdkjebaaxjwnqwzc3j8kvwpoxhup8qck6ud6qkugowuloyidbrm7mf+fdukib/ox5gcfaqrpc5cccq4qxtmopszbzdqqkrm+k6ytovb/r9lbyk9+ert6f43bkthpzxrrebsdvre5+gcxseb3aawekrt1eeuzdgxutrkxodzmctmdvlqgqueu/qrzlnvmmw6qw==</latexit> <latexit sha1_base64="fuufxqu7qwhdpqqluc6c2tk1cla=">aaachhicbzdlsgnbeevrfbtfuzdugoogmzcjgm5ewy1lbancjgw9nrrt2voguymw3yig3/fjqtf3lgq/as/wu4i+lzqclhvrxxdmfpskou+oupdi6nj4xotlanpmdm56vzciulzlbahupxqs5abvdlbbklsejzp5hgo8ds83o/vt69qg5kmx9tnsbxz8rguncyvldd8cpnrefnxjpkil2xx+xtb4khf+wqu8z2woghecmyuaiqnbfu9sx+gvcjtd136oswql747vtkmsykfdem6bkztyregqgwrpi5wyyls36otysjj9giv5xjvuxtptfqbyvidz3v8updamg4e2m+bumb9rpfo/wjonaltvyctlcrmxwbtlilhkekmxttqoshutckgl/ssthw7tiptnxybg/t75l5ys1z237h1t1vy2b2nabczbmqycb1uwbwdwcaqcan38acpzq1z7zw5z4pwiedzzhf+yhn9apyop8=</latexit> <latexit sha1_base64="jm5cs+fc/ppemhhdx4amq8ukv7q=">aaachhicbzdjsgnbeiz7xgpcoh69nazbl2fgbbighepecwcmtddgpmjzl3tvigozbvpgqxjwo4swd4nvywubn/khh468qquv3eyk2vaxnte/slixfgprq6tb2ywtrbrok4vhxqpzayaptmgrqq1fcihmshgos+h4fcvh/xghsgt4uggbwmq3ybiubwhsbyssduobjp3iqpfezs+/yhxewcmq+xh9ih9jxmrh9qyl2evyrbfxskogvobmpkoqpx+na7mu9dijblpnxlsrnsz8m1xejedfmncen9dgstgxelqbezxe53tdohwaxmi9conj/t2qs1hoq+qyzznjv7wh+v+tlwjw1s5elkqier8vcljjmabdpghhkoaobwyyv8l8lfium2mhybnoqncmt56f+lhfssvo9un54mqsr4hskj1yqbxysi7ifamsguhkgtyrf/jqpvrp1pv1pm6dsyyzo+sprm9vybghca==</latexit> <latexit sha1_base64="fuufxqu7qwhdpqqluc6c2tk1cla=">aaachhicbzdlsgnbeevrfbtfuzdugoogmzcjgm5ewy1lbancjgw9nrrt2voguymw3yig3/fjqtf3lgq/as/wu4i+lzqclhvrxxdmfpskou+oupdi6nj4xotlanpmdm56vzciulzlbahupxqs5abvdlbbklsejzp5hgo8ds83o/vt69qg5kmx9tnsbxz8rguncyvldd8cpnrefnxjpkil2xx+xtb4khf+wqu8z2woghecmyuaiqnbfu9sx+gvcjtd136oswql747vtkmsykfdem6bkztyregqgwrpi5wyyls36otysjj9giv5xjvuxtptfqbyvidz3v8updamg4e2m+bumb9rpfo/wjonaltvyctlcrmxwbtlilhkekmxttqoshutckgl/ssthw7tiptnxybg/t75l5ys1z237h1t1vy2b2nabczbmqycb1uwbwdwcaqcan38acpzq1z7zw5z4pwiedzzhf+yhn9apyop8=</latexit> <latexit sha1_base64="fuufxqu7qwhdpqqluc6c2tk1cla=">aaachhicbzdlsgnbeevrfbtfuzdugoogmzcjgm5ewy1lbancjgw9nrrt2voguymw3yig3/fjqtf3lgq/as/wu4i+lzqclhvrxxdmfpskou+oupdi6nj4xotlanpmdm56vzciulzlbahupxqs5abvdlbbklsejzp5hgo8ds83o/vt69qg5kmx9tnsbxz8rguncyvldd8cpnrefnxjpkil2xx+xtb4khf+wqu8z2woghecmyuaiqnbfu9sx+gvcjtd136oswql747vtkmsykfdem6bkztyregqgwrpi5wyyls36otysjj9giv5xjvuxtptfqbyvidz3v8updamg4e2m+bumb9rpfo/wjonaltvyctlcrmxwbtlilhkekmxttqoshutckgl/ssthw7tiptnxybg/t75l5ys1z237h1t1vy2b2nabczbmqycb1uwbwdwcaqcan38acpzq1z7zw5z4pwiedzzhf+yhn9apyop8=</latexit> <latexit sha1_base64="fuufxqu7qwhdpqqluc6c2tk1cla=">aaachhicbzdlsgnbeevrfbtfuzdugoogmzcjgm5ewy1lbancjgw9nrrt2voguymw3yig3/fjqtf3lgq/as/wu4i+lzqclhvrxxdmfpskou+oupdi6nj4xotlanpmdm56vzciulzlbahupxqs5abvdlbbklsejzp5hgo8ds83o/vt69qg5kmx9tnsbxz8rguncyvldd8cpnrefnxjpkil2xx+xtb4khf+wqu8z2woghecmyuaiqnbfu9sx+gvcjtd136oswql747vtkmsykfdem6bkztyregqgwrpi5wyyls36otysjj9giv5xjvuxtptfqbyvidz3v8updamg4e2m+bumb9rpfo/wjonaltvyctlcrmxwbtlilhkekmxttqoshutckgl/ssthw7tiptnxybg/t75l5ys1z237h1t1vy2b2nabczbmqycb1uwbwdwcaqcan38acpzq1z7zw5z4pwiedzzhf+yhn9apyop8=</latexit> Sensi.vi.es with Respect to Ini.al Condi.ons? If an ini.al condi.on is unknown, we could include it as one of the quan..es to be es.mated using least squares Everything goes as before, except that sensi.vity equa.ons are a liele different if you are looking at sensi.vi.es with respect to an ini.al condi.on j () = e j We don t have that second term on the right side because the equa.ons right hand sides of our model don t involve ini.al condi.ons Unit vector with 1 in j th place, elsewhere Deriva.ve of ini.al condi.on i with respect to ini.al condi.on i is one, zero with respect to other ini.al condi.ons So, we now know how to calculate sensi.vi.es with respect to either parameters or ini.al condi.ons; we can calculate both together

26 Sensi.vity Equa.ons for Logis.c Growth Model Because we have an analy.c solu.on of the logis.c growth model dy = ry y 1 ; y() = y K K y(t) = 1+(K/y 1) e rt We can calculate the sensi.vi.es of y(t) with respect to K, r and y using calculus But we could also calculate those three sensi.vi.es using the sensi.vity equa.ons method... and check that the two methods give the same answer! Next slide outlines one of those calcula.ons,

27 Sensi.vity Equa.ons for Logis.c Growth Model Let s set up sensi.vity equa.on transla.ng from general nota.on General Nota.on Nota.on of our problem dx = f(x, t; ) x is y, and θ is r, f is ry(1- y/k) dy = ry 1 y K = + d Calcula.ng f/ y and f/ r, and subs.tu.ng: = r + y 1 y K Have to integrate a two dimensional system, with states y and value of sensi.vity with ini.al

28 Deriva.on of the Sensi.vity Equa.ons for the SIR Model (Appendix of Capaldi et al. 212) We have ds = βsi N di = βsi N γi The four sensi.vi.es, arranged in a matrix: " x θ = # S β I β S γ I γ % ' ' ' & Jacobian: f x " = J(S, I) = # S S ( ds ) I ( ds ) ( di ) I ( di ) % ' " = ' & # βi N βi N βs N βs N γ % ' & Deriva.ve of right hand sides of differen.al equa.ons with respect to parameters " f θ = # β β ds ds ( ) γ ( ) ( ) ( ) di γ di % ' " ' ' = SI N # SI N I & % ' &

29 Deriva.on of the Sensi.vity Equa.ons for the SIR Model We have (Appendix of Capaldi et al. 212) The four sensi.vi.es, arranged in a matrix: Jacobian: ds = βsi N di = βsi N γi f x " = J(S, I) = # S S ( ds ) I ( ds ) ( di ) I ( di ) " x θ = # % ' " = ' & # S β I β βi N βi N S γ I γ % ' ' ' & Shorthand nota.on for sensi.vi.es βs N βs N γ % ' & Deriva.ve of right hand sides of differen.al equa.ons with respect to parameters " f θ = # β β ds ds ( ) γ ( ) ( ) ( ) di γ di % ' " ' ' = SI N # SI N I & % ' &

Sensi&vity Analysis and Least Squares Parameter Es&ma&on for an Epidemic Model. Alun L. Lloyd

Sensi&vity Analysis and Least Squares Parameter Es&ma&on for an Epidemic Model. Alun L. Lloyd Sensi&vity Analysis and Least Squares Parameter Es&ma&on for an Epidemic Model Alun L. Lloyd Department of Mathema&cs Biomathema&cs Graduate Program Center for Quan&ta&ve Sciences in Biomedicine North

More information

Modeling radiocarbon in the Earth System. Radiocarbon summer school 2012

Modeling radiocarbon in the Earth System. Radiocarbon summer school 2012 Modeling radiocarbon in the Earth System Radiocarbon summer school 2012 Outline Brief introduc9on to mathema9cal modeling Single pool models Mul9ple pool models Model implementa9on Parameter es9ma9on What

More information

Numerical Methods in Physics

Numerical Methods in Physics Numerical Methods in Physics Numerische Methoden in der Physik, 515.421. Instructor: Ass. Prof. Dr. Lilia Boeri Room: PH 03 090 Tel: +43-316- 873 8191 Email Address: l.boeri@tugraz.at Room: TDK Seminarraum

More information

Announcements. Topics: Work On: - sec0ons 1.2 and 1.3 * Read these sec0ons and study solved examples in your textbook!

Announcements. Topics: Work On: - sec0ons 1.2 and 1.3 * Read these sec0ons and study solved examples in your textbook! Announcements Topics: - sec0ons 1.2 and 1.3 * Read these sec0ons and study solved examples in your textbook! Work On: - Prac0ce problems from the textbook and assignments from the coursepack as assigned

More information

Ensemble of Climate Models

Ensemble of Climate Models Ensemble of Climate Models Claudia Tebaldi Climate Central and Department of Sta7s7cs, UBC Reto Knu>, Reinhard Furrer, Richard Smith, Bruno Sanso Outline Mul7 model ensembles (MMEs) a descrip7on at face

More information

Ensemble Data Assimila.on and Uncertainty Quan.fica.on

Ensemble Data Assimila.on and Uncertainty Quan.fica.on Ensemble Data Assimila.on and Uncertainty Quan.fica.on Jeffrey Anderson, Alicia Karspeck, Tim Hoar, Nancy Collins, Kevin Raeder, Steve Yeager Na.onal Center for Atmospheric Research Ocean Sciences Mee.ng

More information

Linear Regression and Correla/on. Correla/on and Regression Analysis. Three Ques/ons 9/14/14. Chapter 13. Dr. Richard Jerz

Linear Regression and Correla/on. Correla/on and Regression Analysis. Three Ques/ons 9/14/14. Chapter 13. Dr. Richard Jerz Linear Regression and Correla/on Chapter 13 Dr. Richard Jerz 1 Correla/on and Regression Analysis Correla/on Analysis is the study of the rela/onship between variables. It is also defined as group of techniques

More information

Linear Regression and Correla/on

Linear Regression and Correla/on Linear Regression and Correla/on Chapter 13 Dr. Richard Jerz 1 Correla/on and Regression Analysis Correla/on Analysis is the study of the rela/onship between variables. It is also defined as group of techniques

More information

Bias/variance tradeoff, Model assessment and selec+on

Bias/variance tradeoff, Model assessment and selec+on Applied induc+ve learning Bias/variance tradeoff, Model assessment and selec+on Pierre Geurts Department of Electrical Engineering and Computer Science University of Liège October 29, 2012 1 Supervised

More information

Least Mean Squares Regression. Machine Learning Fall 2017

Least Mean Squares Regression. Machine Learning Fall 2017 Least Mean Squares Regression Machine Learning Fall 2017 1 Lecture Overview Linear classifiers What func?ons do linear classifiers express? Least Squares Method for Regression 2 Where are we? Linear classifiers

More information

Tangent lines, cont d. Linear approxima5on and Newton s Method

Tangent lines, cont d. Linear approxima5on and Newton s Method Tangent lines, cont d Linear approxima5on and Newton s Method Last 5me: A challenging tangent line problem, because we had to figure out the point of tangency.?? (A) I get it! (B) I think I see how we

More information

Structural Equa+on Models: The General Case. STA431: Spring 2013

Structural Equa+on Models: The General Case. STA431: Spring 2013 Structural Equa+on Models: The General Case STA431: Spring 2013 An Extension of Mul+ple Regression More than one regression- like equa+on Includes latent variables Variables can be explanatory in one equa+on

More information

DART Tutorial Sec'on 1: Filtering For a One Variable System

DART Tutorial Sec'on 1: Filtering For a One Variable System DART Tutorial Sec'on 1: Filtering For a One Variable System UCAR The Na'onal Center for Atmospheric Research is sponsored by the Na'onal Science Founda'on. Any opinions, findings and conclusions or recommenda'ons

More information

Pseudospectral Methods For Op2mal Control. Jus2n Ruths March 27, 2009

Pseudospectral Methods For Op2mal Control. Jus2n Ruths March 27, 2009 Pseudospectral Methods For Op2mal Control Jus2n Ruths March 27, 2009 Introduc2on Pseudospectral methods arose to find solu2ons to Par2al Differen2al Equa2ons Recently adapted for Op2mal Control Key Ideas

More information

Some Review and Hypothesis Tes4ng. Friday, March 15, 13

Some Review and Hypothesis Tes4ng. Friday, March 15, 13 Some Review and Hypothesis Tes4ng Outline Discussing the homework ques4ons from Joey and Phoebe Review of Sta4s4cal Inference Proper4es of OLS under the normality assump4on Confidence Intervals, T test,

More information

Introduction to Particle Filters for Data Assimilation

Introduction to Particle Filters for Data Assimilation Introduction to Particle Filters for Data Assimilation Mike Dowd Dept of Mathematics & Statistics (and Dept of Oceanography Dalhousie University, Halifax, Canada STATMOS Summer School in Data Assimila5on,

More information

Quan&fying Uncertainty. Sai Ravela Massachuse7s Ins&tute of Technology

Quan&fying Uncertainty. Sai Ravela Massachuse7s Ins&tute of Technology Quan&fying Uncertainty Sai Ravela Massachuse7s Ins&tute of Technology 1 the many sources of uncertainty! 2 Two days ago 3 Quan&fying Indefinite Delay 4 Finally 5 Quan&fying Indefinite Delay P(X=delay M=

More information

CS 6140: Machine Learning Spring What We Learned Last Week 2/26/16

CS 6140: Machine Learning Spring What We Learned Last Week 2/26/16 Logis@cs CS 6140: Machine Learning Spring 2016 Instructor: Lu Wang College of Computer and Informa@on Science Northeastern University Webpage: www.ccs.neu.edu/home/luwang Email: luwang@ccs.neu.edu Sign

More information

1. Introduc9on 2. Bivariate Data 3. Linear Analysis of Data

1. Introduc9on 2. Bivariate Data 3. Linear Analysis of Data Lecture 3: Bivariate Data & Linear Regression 1. Introduc9on 2. Bivariate Data 3. Linear Analysis of Data a) Freehand Linear Fit b) Least Squares Fit c) Interpola9on/Extrapola9on 4. Correla9on 1. Introduc9on

More information

Experimental Designs for Planning Efficient Accelerated Life Tests

Experimental Designs for Planning Efficient Accelerated Life Tests Experimental Designs for Planning Efficient Accelerated Life Tests Kangwon Seo and Rong Pan School of Compu@ng, Informa@cs, and Decision Systems Engineering Arizona State University ASTR 2015, Sep 9-11,

More information

Data Processing Techniques

Data Processing Techniques Universitas Gadjah Mada Department of Civil and Environmental Engineering Master in Engineering in Natural Disaster Management Data Processing Techniques Hypothesis Tes,ng 1 Hypothesis Testing Mathema,cal

More information

Least Square Es?ma?on, Filtering, and Predic?on: ECE 5/639 Sta?s?cal Signal Processing II: Linear Es?ma?on

Least Square Es?ma?on, Filtering, and Predic?on: ECE 5/639 Sta?s?cal Signal Processing II: Linear Es?ma?on Least Square Es?ma?on, Filtering, and Predic?on: Sta?s?cal Signal Processing II: Linear Es?ma?on Eric Wan, Ph.D. Fall 2015 1 Mo?va?ons If the second-order sta?s?cs are known, the op?mum es?mator is given

More information

Regression Part II. One- factor ANOVA Another dummy variable coding scheme Contrasts Mul?ple comparisons Interac?ons

Regression Part II. One- factor ANOVA Another dummy variable coding scheme Contrasts Mul?ple comparisons Interac?ons Regression Part II One- factor ANOVA Another dummy variable coding scheme Contrasts Mul?ple comparisons Interac?ons One- factor Analysis of variance Categorical Explanatory variable Quan?ta?ve Response

More information

Linear Regression with mul2ple variables. Mul2ple features. Machine Learning

Linear Regression with mul2ple variables. Mul2ple features. Machine Learning Linear Regression with mul2ple variables Mul2ple features Machine Learning Mul4ple features (variables). Size (feet 2 ) Price ($1000) 2104 460 1416 232 1534 315 852 178 Mul4ple features (variables). Size

More information

Structural Equa+on Models: The General Case. STA431: Spring 2015

Structural Equa+on Models: The General Case. STA431: Spring 2015 Structural Equa+on Models: The General Case STA431: Spring 2015 An Extension of Mul+ple Regression More than one regression- like equa+on Includes latent variables Variables can be explanatory in one equa+on

More information

Bellman s Curse of Dimensionality

Bellman s Curse of Dimensionality Bellman s Curse of Dimensionality n- dimensional state space Number of states grows exponen

More information

CSE446: Linear Regression Regulariza5on Bias / Variance Tradeoff Winter 2015

CSE446: Linear Regression Regulariza5on Bias / Variance Tradeoff Winter 2015 CSE446: Linear Regression Regulariza5on Bias / Variance Tradeoff Winter 2015 Luke ZeElemoyer Slides adapted from Carlos Guestrin Predic5on of con5nuous variables Billionaire says: Wait, that s not what

More information

7. Quantum Monte Carlo (QMC)

7. Quantum Monte Carlo (QMC) Molecular Simulations with Chemical and Biological Applications (Part I) 7. Quantum Monte Carlo (QMC) Dr. Mar(n Steinhauser 1 HS 2014 Molecular Simula(ons with Chemical and Biological Applica(ons 1 Introduc5on

More information

Introduc)on to MATLAB for Control Engineers. EE 447 Autumn 2008 Eric Klavins

Introduc)on to MATLAB for Control Engineers. EE 447 Autumn 2008 Eric Klavins Introduc)on to MATLAB for Control Engineers EE 447 Autumn 28 Eric Klavins Part I Matrices and Vectors Whitespace separates entries and a semicolon separates rows. >> A=[1 2 3; 4 5 6; 7 8 9]; # A 3x3 matrix

More information

Garvan Ins)tute Biosta)s)cal Workshop 16/6/2015. Tuan V. Nguyen. Garvan Ins)tute of Medical Research Sydney, Australia

Garvan Ins)tute Biosta)s)cal Workshop 16/6/2015. Tuan V. Nguyen. Garvan Ins)tute of Medical Research Sydney, Australia Garvan Ins)tute Biosta)s)cal Workshop 16/6/2015 Tuan V. Nguyen Tuan V. Nguyen Garvan Ins)tute of Medical Research Sydney, Australia Introduction to linear regression analysis Purposes Ideas of regression

More information

Short introduc,on to the

Short introduc,on to the OXFORD NEUROIMAGING PRIMERS Short introduc,on to the An General Introduction Linear Model to Neuroimaging for Neuroimaging Analysis Mark Jenkinson Mark Jenkinson Janine Michael Bijsterbosch Chappell Michael

More information

UVA CS 4501: Machine Learning. Lecture 6: Linear Regression Model with Dr. Yanjun Qi. University of Virginia

UVA CS 4501: Machine Learning. Lecture 6: Linear Regression Model with Dr. Yanjun Qi. University of Virginia UVA CS 4501: Machine Learning Lecture 6: Linear Regression Model with Regulariza@ons Dr. Yanjun Qi University of Virginia Department of Computer Science Where are we? è Five major sec@ons of this course

More information

Machine Learning and Data Mining. Linear regression. Prof. Alexander Ihler

Machine Learning and Data Mining. Linear regression. Prof. Alexander Ihler + Machine Learning and Data Mining Linear regression Prof. Alexander Ihler Supervised learning Notation Features x Targets y Predictions ŷ Parameters θ Learning algorithm Program ( Learner ) Change µ Improve

More information

CS 6140: Machine Learning Spring What We Learned Last Week. Survey 2/26/16. VS. Model

CS 6140: Machine Learning Spring What We Learned Last Week. Survey 2/26/16. VS. Model Logis@cs CS 6140: Machine Learning Spring 2016 Instructor: Lu Wang College of Computer and Informa@on Science Northeastern University Webpage: www.ccs.neu.edu/home/luwang Email: luwang@ccs.neu.edu Assignment

More information

CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on

CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on Professor Wei-Min Shen Week 13.1 and 13.2 1 Status Check Extra credits? Announcement Evalua/on process will start soon

More information

CS 6140: Machine Learning Spring 2016

CS 6140: Machine Learning Spring 2016 CS 6140: Machine Learning Spring 2016 Instructor: Lu Wang College of Computer and Informa?on Science Northeastern University Webpage: www.ccs.neu.edu/home/luwang Email: luwang@ccs.neu.edu Logis?cs Assignment

More information

REGRESSION AND CORRELATION ANALYSIS

REGRESSION AND CORRELATION ANALYSIS Problem 1 Problem 2 A group of 625 students has a mean age of 15.8 years with a standard devia>on of 0.6 years. The ages are normally distributed. How many students are younger than 16.2 years? REGRESSION

More information

Applied Time Series Analysis FISH 507. Eric Ward Mark Scheuerell Eli Holmes

Applied Time Series Analysis FISH 507. Eric Ward Mark Scheuerell Eli Holmes Applied Time Series Analysis FISH 507 Eric Ward Mark Scheuerell Eli Holmes Introduc;ons Who are we? Who & why you re here? What are you looking to get from this class? Days and Times Lectures When: Tues

More information

Introduc)on to the Design and Analysis of Experiments. Violet R. Syro)uk School of Compu)ng, Informa)cs, and Decision Systems Engineering

Introduc)on to the Design and Analysis of Experiments. Violet R. Syro)uk School of Compu)ng, Informa)cs, and Decision Systems Engineering Introduc)on to the Design and Analysis of Experiments Violet R. Syro)uk School of Compu)ng, Informa)cs, and Decision Systems Engineering 1 Complex Engineered Systems What makes an engineered system complex?

More information

Nonlinear ensemble data assimila/on in high- dimensional spaces. Peter Jan van Leeuwen Javier Amezcua, Mengbin Zhu, Melanie Ades

Nonlinear ensemble data assimila/on in high- dimensional spaces. Peter Jan van Leeuwen Javier Amezcua, Mengbin Zhu, Melanie Ades Nonlinear ensemble data assimila/on in high- dimensional spaces Peter Jan van Leeuwen Javier Amezcua, Mengbin Zhu, Melanie Ades Data assimila/on: general formula/on Bayes theorem: The solu/on is a pdf!

More information

Last Lecture Recap UVA CS / Introduc8on to Machine Learning and Data Mining. Lecture 3: Linear Regression

Last Lecture Recap UVA CS / Introduc8on to Machine Learning and Data Mining. Lecture 3: Linear Regression UVA CS 4501-001 / 6501 007 Introduc8on to Machine Learning and Data Mining Lecture 3: Linear Regression Yanjun Qi / Jane University of Virginia Department of Computer Science 1 Last Lecture Recap q Data

More information

Class Notes. Examining Repeated Measures Data on Individuals

Class Notes. Examining Repeated Measures Data on Individuals Ronald Heck Week 12: Class Notes 1 Class Notes Examining Repeated Measures Data on Individuals Generalized linear mixed models (GLMM) also provide a means of incorporang longitudinal designs with categorical

More information

Exponen'al growth and differen'al equa'ons

Exponen'al growth and differen'al equa'ons Exponen'al growth and differen'al equa'ons But first.. Thanks for the feedback! Feedback about M102 Which of the following do you find useful? 70 60 50 40 30 20 10 0 How many resources students typically

More information

Machine learning for Dynamic Social Network Analysis

Machine learning for Dynamic Social Network Analysis Machine learning for Dynamic Social Network Analysis Manuel Gomez Rodriguez Max Planck Ins7tute for So;ware Systems UC3M, MAY 2017 Interconnected World SOCIAL NETWORKS TRANSPORTATION NETWORKS WORLD WIDE

More information

CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on. Professor Wei-Min Shen Week 8.1 and 8.2

CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on. Professor Wei-Min Shen Week 8.1 and 8.2 CSCI 360 Introduc/on to Ar/ficial Intelligence Week 2: Problem Solving and Op/miza/on Professor Wei-Min Shen Week 8.1 and 8.2 Status Check Projects Project 2 Midterm is coming, please do your homework!

More information

Comments on Black Hole Interiors

Comments on Black Hole Interiors Comments on Black Hole Interiors Juan Maldacena Ins6tute for Advanced Study Conserva6ve point of view Expansion parameter = geff 2 ld 2 p r D 2 h 1 S Informa6on seems to be lost perturba6vely. n point

More information

Announcements. Topics: Homework: - sec0ons 1.2, 1.3, and 2.1 * Read these sec0ons and study solved examples in your textbook!

Announcements. Topics: Homework: - sec0ons 1.2, 1.3, and 2.1 * Read these sec0ons and study solved examples in your textbook! Announcements Topics: - sec0ons 1.2, 1.3, and 2.1 * Read these sec0ons and study solved examples in your textbook! Homework: - review lecture notes thoroughly - work on prac0ce problems from the textbook

More information

Logis&c Regression. Robot Image Credit: Viktoriya Sukhanova 123RF.com

Logis&c Regression. Robot Image Credit: Viktoriya Sukhanova 123RF.com Logis&c Regression These slides were assembled by Eric Eaton, with grateful acknowledgement of the many others who made their course materials freely available online. Feel free to reuse or adapt these

More information

Two common difficul:es with HW 2: Problem 1c: v = r n ˆr

Two common difficul:es with HW 2: Problem 1c: v = r n ˆr Two common difficul:es with HW : Problem 1c: For what values of n does the divergence of v = r n ˆr diverge at the origin? In this context, diverge means becomes arbitrarily large ( goes to infinity ).

More information

CSE 473: Ar+ficial Intelligence

CSE 473: Ar+ficial Intelligence CSE 473: Ar+ficial Intelligence Hidden Markov Models Luke Ze@lemoyer - University of Washington [These slides were created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188

More information

Modelling Time- varying effec3ve Brain Connec3vity using Mul3regression Dynamic Models. Thomas Nichols University of Warwick

Modelling Time- varying effec3ve Brain Connec3vity using Mul3regression Dynamic Models. Thomas Nichols University of Warwick Modelling Time- varying effec3ve Brain Connec3vity using Mul3regression Dynamic Models Thomas Nichols University of Warwick Dynamic Linear Model Bayesian 3me series model Predictors {X 1,,X p } Exogenous

More information

Ensemble Data Assimila.on for Climate System Component Models

Ensemble Data Assimila.on for Climate System Component Models Ensemble Data Assimila.on for Climate System Component Models Jeffrey Anderson Na.onal Center for Atmospheric Research In collabora.on with: Alicia Karspeck, Kevin Raeder, Tim Hoar, Nancy Collins IMA 11

More information

Extreme value sta-s-cs of smooth Gaussian random fields in cosmology

Extreme value sta-s-cs of smooth Gaussian random fields in cosmology Extreme value sta-s-cs of smooth Gaussian random fields in cosmology Stéphane Colombi Ins2tut d Astrophysique de Paris with O. Davis, J. Devriendt, S. Prunet, J. Silk The large scale galaxy distribu-on

More information

Engineering Characteriza.on of Spa.ally Variable Ground Mo.on

Engineering Characteriza.on of Spa.ally Variable Ground Mo.on Engineering Characteriza.on of Spa.ally Variable Ground Mo.on Timothy D. Ancheta PEER Center, UC Berkeley Jonathan P. Stewart UCLA Civil & Environmental Engineering Department Norman A. Abrahamson Pacific

More information

CSE 473: Ar+ficial Intelligence. Hidden Markov Models. Bayes Nets. Two random variable at each +me step Hidden state, X i Observa+on, E i

CSE 473: Ar+ficial Intelligence. Hidden Markov Models. Bayes Nets. Two random variable at each +me step Hidden state, X i Observa+on, E i CSE 473: Ar+ficial Intelligence Bayes Nets Daniel Weld [Most slides were created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188 materials are available at hnp://ai.berkeley.edu.]

More information

Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval. Lecture #3 Machine Learning. Edward Chang

Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval. Lecture #3 Machine Learning. Edward Chang Founda'ons of Large- Scale Mul'media Informa'on Management and Retrieval Lecture #3 Machine Learning Edward Y. Chang Edward Chang Founda'ons of LSMM 1 Edward Chang Foundations of LSMM 2 Machine Learning

More information

Reduced Models for Process Simula2on and Op2miza2on

Reduced Models for Process Simula2on and Op2miza2on Reduced Models for Process Simulaon and Opmizaon Yidong Lang, Lorenz T. Biegler and David Miller ESI annual meeng March, 0 Models are mapping Equaon set or Module simulators Input space Reduced model Surrogate

More information

Lecture 12 The Level Set Approach for Turbulent Premixed Combus=on

Lecture 12 The Level Set Approach for Turbulent Premixed Combus=on Lecture 12 The Level Set Approach for Turbulent Premixed Combus=on 12.- 1 A model for premixed turbulent combus7on, based on the non- reac7ng scalar G rather than on progress variable, has been developed

More information

COMP 562: Introduction to Machine Learning

COMP 562: Introduction to Machine Learning COMP 562: Introduction to Machine Learning Lecture 20 : Support Vector Machines, Kernels Mahmoud Mostapha 1 Department of Computer Science University of North Carolina at Chapel Hill mahmoudm@cs.unc.edu

More information

Par$cle Filters Part I: Theory. Peter Jan van Leeuwen Data- Assimila$on Research Centre DARC University of Reading

Par$cle Filters Part I: Theory. Peter Jan van Leeuwen Data- Assimila$on Research Centre DARC University of Reading Par$cle Filters Part I: Theory Peter Jan van Leeuwen Data- Assimila$on Research Centre DARC University of Reading Reading July 2013 Why Data Assimila$on Predic$on Model improvement: - Parameter es$ma$on

More information

The Mysteries of Quantum Mechanics

The Mysteries of Quantum Mechanics The Mysteries of Quantum Mechanics Class 5: Quantum Behavior and Interpreta=ons Steve Bryson www.stevepur.com/quantum Ques=ons? The Quantum Wave Quantum Mechanics says: A par=cle s behavior is described

More information

MA/CS 109 Lecture 7. Back To Exponen:al Growth Popula:on Models

MA/CS 109 Lecture 7. Back To Exponen:al Growth Popula:on Models MA/CS 109 Lecture 7 Back To Exponen:al Growth Popula:on Models Homework this week 1. Due next Thursday (not Tuesday) 2. Do most of computa:ons in discussion next week 3. If possible, bring your laptop

More information

Introduc)on to Ar)ficial Intelligence

Introduc)on to Ar)ficial Intelligence Introduc)on to Ar)ficial Intelligence Lecture 10 Probability CS/CNS/EE 154 Andreas Krause Announcements! Milestone due Nov 3. Please submit code to TAs! Grading: PacMan! Compiles?! Correct? (Will clear

More information

Sample sta*s*cs and linear regression. NEU 466M Instructor: Professor Ila R. Fiete Spring 2016

Sample sta*s*cs and linear regression. NEU 466M Instructor: Professor Ila R. Fiete Spring 2016 Sample sta*s*cs and linear regression NEU 466M Instructor: Professor Ila R. Fiete Spring 2016 Mean {x 1,,x N } N samples of variable x hxi 1 N NX i=1 x i sample mean mean(x) other notation: x Binned version

More information

Miscellaneous Thoughts on Ocean Data Assimilation

Miscellaneous Thoughts on Ocean Data Assimilation Miscellaneous Thoughts on Ocean Data Assimilation Mike Dowd Dept of Mathematics & Statistics (and Dept of Oceanography) Dalhousie University, Halifax, Canada STATMOS Summer School in Data Assimila5on,

More information

ODEs + Singulari0es + Monodromies + Boundary condi0ons. Kerr BH ScaRering: a systema0c study. Schwarzschild BH ScaRering: Quasi- normal modes

ODEs + Singulari0es + Monodromies + Boundary condi0ons. Kerr BH ScaRering: a systema0c study. Schwarzschild BH ScaRering: Quasi- normal modes Outline Introduc0on Overview of the Technique ODEs + Singulari0es + Monodromies + Boundary condi0ons Results Kerr BH ScaRering: a systema0c study Schwarzschild BH ScaRering: Quasi- normal modes Collabora0on:

More information

Regression.

Regression. Regression www.biostat.wisc.edu/~dpage/cs760/ Goals for the lecture you should understand the following concepts linear regression RMSE, MAE, and R-square logistic regression convex functions and sets

More information

STAD68: Machine Learning

STAD68: Machine Learning STAD68: Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! h0p://www.cs.toronto.edu/~rsalakhu/ Lecture 1 Evalua;on 3 Assignments worth 40%. Midterm worth 20%. Final

More information

MATH 5740/MATH MATH MODELING :40 AM- 10:30 AM JWB 208 Andrej Cherkaev. Introduc*on

MATH 5740/MATH MATH MODELING :40 AM- 10:30 AM JWB 208 Andrej Cherkaev. Introduc*on MATH 5740/MATH 6870 001 MATH MODELING 2013 09:40 AM- 10:30 AM JWB 208 Andrej Cherkaev Introduc*on Components of applied math Math Modeling Differen*al equa*ons Numerical Methods Data: Sta*s*cs approx Applied

More information

Next is material on matrix rank. Please see the handout

Next is material on matrix rank. Please see the handout B90.330 / C.005 NOTES for Wednesday 0.APR.7 Suppose that the model is β + ε, but ε does not have the desired variance matrix. Say that ε is normal, but Var(ε) σ W. The form of W is W w 0 0 0 0 0 0 w 0

More information

Machine Learning and Data Mining. Bayes Classifiers. Prof. Alexander Ihler

Machine Learning and Data Mining. Bayes Classifiers. Prof. Alexander Ihler + Machine Learning and Data Mining Bayes Classifiers Prof. Alexander Ihler A basic classifier Training data D={x (i),y (i) }, Classifier f(x ; D) Discrete feature vector x f(x ; D) is a con@ngency table

More information

Outline. What is Machine Learning? Why Machine Learning? 9/29/08. Machine Learning Approaches to Biological Research: Bioimage Informa>cs and Beyond

Outline. What is Machine Learning? Why Machine Learning? 9/29/08. Machine Learning Approaches to Biological Research: Bioimage Informa>cs and Beyond Outline Machine Learning Approaches to Biological Research: Bioimage Informa>cs and Beyond Robert F. Murphy External Senior Fellow, Freiburg Ins>tute for Advanced Studies Ray and Stephanie Lane Professor

More information

Mixture Models. Michael Kuhn

Mixture Models. Michael Kuhn Mixture Models Michael Kuhn 2017-8-26 Objec

More information

The Color Glass Condensate: Theory, Experiment and the Future

The Color Glass Condensate: Theory, Experiment and the Future The Color Glass Condensate: Theory, Experiment and the Future Physics Issues: What is the high energy limit of strong interac?ons? How do we compute the gluon and quark distribu?ons relevant for asympto?cally

More information

Parallelizing Gaussian Process Calcula1ons in R

Parallelizing Gaussian Process Calcula1ons in R Parallelizing Gaussian Process Calcula1ons in R Christopher Paciorek UC Berkeley Sta1s1cs Joint work with: Benjamin Lipshitz Wei Zhuo Prabhat Cari Kaufman Rollin Thomas UC Berkeley EECS (formerly) IBM

More information

Building your toolbelt

Building your toolbelt Building your toolbelt Using math to make meaning in the physical world. Dimensional analysis Func;onal dependence / scaling Special cases / limi;ng cases Reading the physics in the representa;on (graphs)

More information

PSAAP Project Stanford

PSAAP Project Stanford PSAAP Project QMU @ Stanford Component Analysis and rela:on to Full System Simula:ons 1 What do we want to predict? Objec:ve: predic:on of the unstart limit expressed as probability of unstart (or alterna:vely

More information

An Introduc+on to Sta+s+cs and Machine Learning for Quan+ta+ve Biology. Anirvan Sengupta Dept. of Physics and Astronomy Rutgers University

An Introduc+on to Sta+s+cs and Machine Learning for Quan+ta+ve Biology. Anirvan Sengupta Dept. of Physics and Astronomy Rutgers University An Introduc+on to Sta+s+cs and Machine Learning for Quan+ta+ve Biology Anirvan Sengupta Dept. of Physics and Astronomy Rutgers University Why Do We Care? Necessity in today s labs Principled approach:

More information

Computer Vision. Pa0ern Recogni4on Concepts Part I. Luis F. Teixeira MAP- i 2012/13

Computer Vision. Pa0ern Recogni4on Concepts Part I. Luis F. Teixeira MAP- i 2012/13 Computer Vision Pa0ern Recogni4on Concepts Part I Luis F. Teixeira MAP- i 2012/13 What is it? Pa0ern Recogni4on Many defini4ons in the literature The assignment of a physical object or event to one of

More information

Some thoughts on linearity, nonlinearity, and partial separability

Some thoughts on linearity, nonlinearity, and partial separability Some thoughts on linearity, nonlinearity, and partial separability Paul Hovland Argonne Na0onal Laboratory Joint work with Boyana Norris, Sri Hari Krishna Narayanan, Jean Utke, Drew Wicke Argonne Na0onal

More information

Regulatory Inferece from Gene Expression. CMSC858P Spring 2012 Hector Corrada Bravo

Regulatory Inferece from Gene Expression. CMSC858P Spring 2012 Hector Corrada Bravo Regulatory Inferece from Gene Expression CMSC858P Spring 2012 Hector Corrada Bravo 2 Graphical Model Let y be a vector- valued random variable Suppose some condi8onal independence proper8es hold for some

More information

Priors in Dependency network learning

Priors in Dependency network learning Priors in Dependency network learning Sushmita Roy sroy@biostat.wisc.edu Computa:onal Network Biology Biosta2s2cs & Medical Informa2cs 826 Computer Sciences 838 hbps://compnetbiocourse.discovery.wisc.edu

More information

Modelling of Equipment, Processes, and Systems

Modelling of Equipment, Processes, and Systems 1 Modelling of Equipment, Processes, and Systems 2 Modelling Tools Simple Programs Spreadsheet tools like Excel Mathema7cal Tools MatLab, Mathcad, Maple, and Mathema7ca Special Purpose Codes Macroflow,

More information

Deriva'on of The Kalman Filter. Fred DePiero CalPoly State University EE 525 Stochas'c Processes

Deriva'on of The Kalman Filter. Fred DePiero CalPoly State University EE 525 Stochas'c Processes Deriva'on of The Kalman Filter Fred DePiero CalPoly State University EE 525 Stochas'c Processes KF Uses State Predic'ons KF es'mates the state of a system Example Measure: posi'on State: [ posi'on velocity

More information

Void abundance predic.on and spherical evolu.on model.!!!!! Ixandra Achitouv! Universitäts- Sternwarte München!

Void abundance predic.on and spherical evolu.on model.!!!!! Ixandra Achitouv! Universitäts- Sternwarte München! Void abundance predic.on and spherical evolu.on model!!!!! Ixandra Achitouv! Universitäts- Sternwarte München! Voids hierarchy - Sheth & Van de Weijgaert (04) Voids Hierarchies: Number of voids dn/dr at

More information

Week 12, Lecture 2 Nuclear Synthesis

Week 12, Lecture 2 Nuclear Synthesis Week 12, Lecture 2 Nuclear Synthesis Nuclear Reac*ons in Space - - Overview - - Observa

More information

STA 4273H: Sta-s-cal Machine Learning

STA 4273H: Sta-s-cal Machine Learning STA 4273H: Sta-s-cal Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistical Sciences! rsalakhu@cs.toronto.edu! h0p://www.cs.utoronto.ca/~rsalakhu/ Lecture 1 Evalua:on

More information

Integra(ng and Ranking Uncertain Scien(fic Data

Integra(ng and Ranking Uncertain Scien(fic Data Jan 19, 2010 1 Biomedical and Health Informatics 2 Computer Science and Engineering University of Washington Integra(ng and Ranking Uncertain Scien(fic Data Wolfgang Ga*erbauer 2 Based on joint work with:

More information

Gradient Descent for High Dimensional Systems

Gradient Descent for High Dimensional Systems Gradient Descent for High Dimensional Systems Lab versus Lab 2 D Geometry Op>miza>on Poten>al Energy Methods: Implemented Equa3ons for op3mizer 3 2 4 Bond length High Dimensional Op>miza>on Applica3ons:

More information

DART Tutorial Sec'on 9: More on Dealing with Error: Infla'on

DART Tutorial Sec'on 9: More on Dealing with Error: Infla'on DART Tutorial Sec'on 9: More on Dealing with Error: Infla'on UCAR The Na'onal Center for Atmospheric Research is sponsored by the Na'onal Science Founda'on. Any opinions, findings and conclusions or recommenda'ons

More information

Sta$s$cs in astronomy and the SAMSI ASTRO Program

Sta$s$cs in astronomy and the SAMSI ASTRO Program Sta$s$cs in astronomy and the SAMSI ASTRO Program G. Jogesh Babu (with input from Eric Feigelson) Penn State Why astrosta$s$cs? Astronomers encounter a surprising variety of sta$s$cal problems in their

More information

Introduction to Statistical Genetics (BST227) Lecture 6: Population Substructure in Association Studies

Introduction to Statistical Genetics (BST227) Lecture 6: Population Substructure in Association Studies Introduction to Statistical Genetics (BST227) Lecture 6: Population Substructure in Association Studies Confounding in gene+c associa+on studies q What is it? q What is the effect? q How to detect it?

More information

One- factor ANOVA. F Ra5o. If H 0 is true. F Distribu5on. If H 1 is true 5/25/12. One- way ANOVA: A supersized independent- samples t- test

One- factor ANOVA. F Ra5o. If H 0 is true. F Distribu5on. If H 1 is true 5/25/12. One- way ANOVA: A supersized independent- samples t- test F Ra5o F = variability between groups variability within groups One- factor ANOVA If H 0 is true random error F = random error " µ F =1 If H 1 is true random error +(treatment effect)2 F = " µ F >1 random

More information

A6523 Modeling, Inference, and Mining Jim Cordes, Cornell University

A6523 Modeling, Inference, and Mining Jim Cordes, Cornell University A6523 Modeling, Inference, and Mining Jim Cordes, Cornell University Lecture 23 Birthday problem comments Construc

More information

Pu#ng the physics into sea ice parameterisa3ons: a case study (melt ponds)

Pu#ng the physics into sea ice parameterisa3ons: a case study (melt ponds) Pu#ng the physics into sea ice parameterisa3ons: a case study (melt ponds) April 1998 July 1998 Ice Sta3on SHEBA. Canadian Coast Guard icebreaker Des Groseilliers. Danny Feltham Centre for Polar Observa3on

More information

Networks. Can (John) Bruce Keck Founda7on Biotechnology Lab Bioinforma7cs Resource

Networks. Can (John) Bruce Keck Founda7on Biotechnology Lab Bioinforma7cs Resource Networks Can (John) Bruce Keck Founda7on Biotechnology Lab Bioinforma7cs Resource Networks in biology Protein-Protein Interaction Network of Yeast Transcriptional regulatory network of E.coli Experimental

More information

Parallel Tempering Algorithm in Monte Carlo Simula5on

Parallel Tempering Algorithm in Monte Carlo Simula5on Parallel Tempering Algorithm in Monte Carlo Simula5on Tony Cheung (CUHK) Kevin Zhao (CUHK) Mentors: Ying Wai Li (ORNL) Markus Eisenbach (ORNL) Kwai Wong (UTK/ORNL) Monte Carlo Algorithms Mo5va5on: Idea:

More information

Gene Regulatory Networks II Computa.onal Genomics Seyoung Kim

Gene Regulatory Networks II Computa.onal Genomics Seyoung Kim Gene Regulatory Networks II 02-710 Computa.onal Genomics Seyoung Kim Goal: Discover Structure and Func;on of Complex systems in the Cell Identify the different regulators and their target genes that are

More information

A mul&scale autocorrela&on func&on for anisotropy studies

A mul&scale autocorrela&on func&on for anisotropy studies A mul&scale autocorrela&on func&on for anisotropy studies Mario Scuderi 1, M. De Domenico, H Lyberis and A. Insolia 1 Department of Physics and Astronomy & INFN Catania University ITALY DAA2011 Erice,

More information