Continuous-time Nonlinear Estimation Filters Using UKF-aided Gaussian Sum Representations

Size: px
Start display at page:

Download "Continuous-time Nonlinear Estimation Filters Using UKF-aided Gaussian Sum Representations"

Transcription

1 Connuous-me onlnear Esmaon Flers Usn UKF-aded Gaussan Sum Reresenaons ura Gokce echnolo and Innovaon Fundn rorams Drecorae he Scenfc and echnolocal Research Councl of urke Ankara urke usafa Kuzuolu Elecrcal and Elecroncs Enneern Dearmen ddle Eas echncal Unvers Ankara urke Absrac An aromae nonlnear esmaon mehod for connuous-me ssems h dscree-me measuremens s develoed. he aroach evaluaes he Gaussan sum aromaon of he a ror robabl dens funcon df b solvn he Fokker-lanck equaon numercall. Aromae evaluaon of he a oseror df s acheved b usn Gaussan sums a ror df and measuremens n Baes rule. ean and covarance values of Gaussans are chosen b he hel of an Unscened Kalman Fler UKF h resec o a reon here a ror and a oseror dfs are aromaed. Wehs of he Gaussans are udaed usn he deermnscall chosen rd ons n he secfed domans. UKF here acs as a one se look ahead mechansm o deermne he hh robabl reons here a ror and a oseror dfs can resde. he a ror and a oseror dfs are aromaed around hese hh robabl reons. he develoed aroach s comared h UKF and arcle Fler n a one dmensonal nonlnear ssem. Keords nonlnear ssem; flern alorhm; connuousme ssem; Fokker-lanck equaon; Gaussan sum; numercal mehods I. IRODUCIO he eneral roblem of nonlnear flern can be handled for ssems ha are reresened eher n connuous-me or dscree-me. Connuous-me ssems are more eneral and can be aromaed b dscree-me ssems usn suable dscrezaon mehods. here es a fe eac.e. omal nonlnear flers n connuous-me for some secal cases and here are several aromae flern mehods aled o boh connuous-me and dscree-me ssems n he leraure. A dnamcal ssem ves he me evoluon of he sae and a measuremen model can be nroduced hch elds samles of he sae a a secfc me nsan. For he connuous-me case e assume a connuous-me dnamcal ssem and a dscree-me measuremen model. he sae s he sochasc rocess hose me evoluon s overned b he Io sochasc dfferenal equaon. d =f d + G dß h he nal condon =. easuremens are obaned n dscree-me as: k =h k k + v k є R n denoes he sae vecor f є R n s he drf funcon and ß є R m s a Wener rocess and he ncremens of Wener rocess dß mus be nfnel dvsble [7]. G є R nm s a sae deenden mar. he ndeenden varable reresens me. denoes he nal me nsan and k s he k-h samln me. he nal sae s assumed o be a random vecor h a knon no necessarl Gaussan dens. є R s a vecor of measuremens h є R s a funcon of he sae and me and v є R s a zero-mean he measuremen nose. I s assumed ha he random varables ß and v are muuall ndeenden. In order o smlf he noaon e ll re k for k ec. he flern roblem nvolves esman he values of he saes of he ssem b he hel of he measuremens. Snce samled saes are random varables e need o esmae her condonal robabl dens funcon ha deends on he measuremens a samln me k hch s called a oseror dens k k. Afer he deermnaon of he a oseror dens he sae esmae ma be comued. For he ven connuous-me dnamcal ssem he condonal robabl dens funcon evolves beeen o measuremens or o samln me nsans b means of he Fokker-lanck equaon eldn he a ror dens hch s udaed h Baes' rule usn measuremens a curren samln me vn a oseror dens. he recursve soluon of he flern roblem s ven n []. he Fokker- lanck equaon s ven as f GQβG f r r he Fokker-lanck equaon s n eneral a nonlnear aral dfferenal equaon hose soluon canno be eressed n erms of a fne number of arameers. Ece for a fe secal cases e.. lnear Gaussan case e canno fnd an analcal soluon of Fokker-lanck equaon for he a ror

2 df hch s used n udae se o fnd he a oseror df. Even f hese dfs are analcall avalable ma be dffcul or mossble o evaluae he nerals for normalzaon uroses or fndn he mean and covarance of he sae. As an alernave aromae mehods are used for solvn he dfferenal equaons or evaluan he nerals. Hoever some of he aromae mehods ma no eld suffcen accurac for ceran cases or ma requre ecessve comuaonal oer o acheve suffcen accurac. Generall he requred comuaonal oer does no ncrease lnearl as he dmenson of he sae varables ncreases. For he connuous-me case he Benes fler [3] and he eended Benes fler Daum fler [3] ve eac soluons for secal cases. In Benes fler he eac soluon of he Fokker- lanck equaon assumes an eonenal dens funcon characerzed n erms of sae and covarance. Daum fler eneralzes he Benes fler for he case here an analcal soluon o Fokker-lanck equaon ess. Eended Kalman Fler seudolnear Fler Second Order Flers and Eanson of Dens and Coordnae ransformaons are some of he aromae mehods for he soluon of roblem. Deals of hese flers are ven n []. As menoned n [] here are o es of aromaons suesed n he leraure eher he model s relaced b a smler one or numercal mehods are used o fnd a lobal aromaon of he a oseror dens. odel aromaon s obaned b eandn he nonlnear funcons around he oeran on a ever me se usn alor seres eansons. hs aroach elds he Eended Kalman Fler EKF. Anoher model aromaon echnque s based on he reresenaon of he sae vecor b a fne se of values. hs model class s referred o as he Hdden arkov odel H. Convered easuremen Kalman Fler CKF s also anoher echnque ha res o lnearze he nonlnear measuremen model b convern he measuremens no he sae sace. hs fler s commonl used n radar rackn alcaons. he oher alernave aroach s based on he aromaon of dfs. For nsance he dfs are aromaed usn a sum of Gaussan denses n []. Anoher df aromaon s rovded b a samln echnque knon as he ransformaon echnque o ck a mnmal se of samle ons called sma ons around he mean. hese sma ons are hen roaaed hrouh he nonlnear funcons from hch he mean and covarance of he esmae are hen recovered. UKF and Quadraure Kalman Fler are of hs form and sma ons are udaed b usn oeraons smlar o hose n he Kalman Fler. UKF s frs ublshed n [6]. An UKF formulaon for he connuous-me case s ven n [4]. Anoher df aromaon knon as he on-mass fler aromaes he a oseror dens b a se of ons on a redefned rd. An eenson of he on-mass fler s he sequenal one Carlo mehod also referred o as arcle Fler. In hs mehod he a oseror dens s also aromaed b a se of ons hoever he rd s chosen n a sochasc raher han n a deermnsc manner. A comarson of some of hese mehods can be found n [5]. he oranzaon of he aer s as follos; In Secon II a flern aroach he UKF-aded Gaussan Sum Fler for connuous-me ssems s elaned. hs mehod s based on he aromaon of dfs b Gaussan sums. Smulaon resuls relaed o hs fler are resened n Secon III. Conclusons are dran n Secon IV. II. UKF AİDED GAUSSIA SU FILER FOR COIUOUS-IE SYSES he aroach elaned n hs secon s a flern mehod based on he aromaon of boh he a ror and he a oseror dfs of he saes of he ven connuous-me ssem usn ehed Gaussan sums and nvolves he follon o ses. Evaluaon of he Gaussan sum aromaon of he a ror df b solvn he Fokker-lanck equaon numercall. Aromae evaluaon of he a oseror df b usn Gaussan sums a ror df and measuremens n Baes rule. ean and covarance values of Gaussans are deermned b he hel of an UKF hrouh hese ses. he deals of hese ses are ven belo. A. Evaluaon of he aromae a ror df For he ven ssem he Fokker-lanck equaon hch s used for fndn he a ror df s ven as follos; f f r GQβG r he rh hand sde of he equaon can be reresened as a nonlnear funcon as; here f f r GQβG r 3 can be solved b usn he Euler s mehod. A e sar h a df. We an o fnd a ror df a denoes he samln me usn 3 and. n=/ Euler ses can be used for hs urose. Hoever e canno aromae he connuous-me

3 bu e can onl evaluae a he samle ons. amel e can fnd for =. In order o fnd an aromae eresson for e emloed a Gaussan sum aromaon. eans covarances and ehs of Gaussans are calculaed a ever se of Euler s mehod. eans and covarances are deermned usn he redcon se of an UKF. Wehs are udaed usn deermnsc rd ons chosen from a reon covern he revous and curren means of he Gaussans. he follon ses summarze mean covarance and eh calculaon n a snle Euler se.. he redcon se of UKF s aled o he mean and covarance values of nal df o fnd he redced mean and covarance values. For hs urose assumn ha he nal df s Gaussan he unscened ransform s aled usn. he nonlnear funcon used n he ransformaon s f here = +. eans of Gaussans are deermned on a unform rd around. he dh of he rd s deermned usn a suabl scaled value of. Covarances of Gaussans are assned o he same value such ha. hs scaln can be done as follos; Under he assumon ha he mean of he sae varable s and for a Gaussan sum aromaon of he a ror df s e here > le he dh of he rd be defned as he mar lda c here l and c s he scaln facor. For c l can be chosen as zero and he covarances of Gaussans can be aken as as here <ν<. For <c< he covarances of Gaussans can be aken as here ν. eans of Gaussans are unforml deermned accordn o he daonal elemens of he dh of he rd. 3. o ses of rd ons are defned. One se nvolves he mean values of Gaussans for ha se. he oher se nvolves ons chosen from a unform rd around. here 4. he a ror df for he curren se s modeled usn Gaussan sums as here e Also e can fnd he values of a ror df a a ven on usn he Euler dscrezed Fokker-lanck Equaon. B equan 5 and 6 a he rd ons e have he follon ssem. B A

4 A. B he leas squares soluon of hs ssem can be ven as A A A B If A s nearl snular mamum values alon each ro of A mar ma be used. In hs case he ar of A + mar ll be daonal for dsnc values avodn he nverson roblems. A he above rocedure s reeaed a ever se. B. Evaluaon of he aromae a oseror df he lkelhood and he aromaed a ror df are combned n Baes rule eldn he unnormalzed a oseror df as; For a nonlnear measuremen model he lkelhood ll no be a smle eresson and he unnormalzed a oseror df hch s calculaed usn he lkelhood canno be used drecl o fnd s mean and covarance. So he unnormalzed a oseror df can be reresened aromael h ehed Gaussans as; K! e ean and covarance values of Gaussans are found n a manner smlar o he a ror df. Here mean and covarance of a ror df are used n he udae se of UKF. Grd ons are chosen as he mean values of Gaussans n hs case. he resuln mar here s an smmerc osve defne mar ven as n [] he reon used o choose he mean values of Gaussans deermnes he search sace here he a ror and a oseror dfs are red o be aromaed. Covarance values mus no be oo hh and number of Gaussans used n he aromaons s also moran for a ood reresenaon. If he number of Gaussans ncreases here can be an overfn. For aromaons nvolvn a fe Gaussans dfs ma no f ell. he number of Gaussans s roblem deenden and ma be chosen emrcall. III. SIULAIO RESULS he develoed aroach s mlemened n a one dmensonal nonlnear ssem and he resuls are comared h hose obaned b solvn he same roblem h he UKF and Sequenal Imorance Samln SIR arcle Fler. Connuous-dscree UKF of [4] s used and also nose free dnamcal model s used n he SIR arcle Fler for me evoluon of arcles before addn nose. he smulaons are carred ou on a comuer h Inel core 3 CU and.8 Gb RA usn ALAB. c&oc command s used for he comuaon me and normalzed accordn o he UKF hle fndn he comuaonal load. Consder he follon one dmensonal connuous-me nonlnear ssem. 5 d 8cos. d d here s a Wener rocess h Gaussan ncremens hose nens s aken as. easuremens are aken b usn he follon dscree me model a a samln erod of =s. k 3 k v here v k s a Gaussan measuremen nose. n= Euler dscrezaon ses beeen o samles are used. Smulaed daa ere eneraed usn he ses Euler- aruama mehod [8] a ever second and smulaon lass for s. For he ven roblem o confuraons are used. he frs one uses Gaussans hle aroman he a ror df and 5 Gaussans hle aroman he a oseror df. he second one uses Gaussans hle aroman he a ror df and Gaussans hle aroman he a oseror df. he resuls of he 3 one Carlo runs are ven n able I n erms of he ean of ean Absolue Error AE and Comuaon Load. AE here =s and =3 runs. ABLE I. Fler o k rue es COARISO OF AE AD COUAIO LOAD Flers ean of ean Absolue Error AE Averae Comuaon Load for One erod Unless UKF SIR arcle Fler h arcles SIR arcle Fler h 5 arcles UKF-aded Gaussan Sum Fler h +5 Gaussans UKF-aded Gaussan Sum Fler h + Gaussans

5 Fure ves he ean of Absolue ErrorAE afer 3 one Carlo runs. 5 AE rue es 5 For he one dmensonal ssem aromael one hundred arcles are enouh for he omal accurac [9]. Bu he number of arcles ncreases eensvel n hher dmensons. For o samle runs he ar of he unnormalzed aromaed dfs n he reon here he means of Gaussans are chosen are ven n F. and F. 3 for he o confuraons a he end of las smulaon me a =s. Almos eac dfs are found usn he hsoram of arcles of a SIR arcle Fler nvolvn arcles. As can be seen from F. and F. 3 for he frs confuraon onl a lmed ar of he df s aromaed n a reon close o he mean. Snce he number of rd ons s small a smaller search reon s chosen. Bu snce aromaon s carred ou n an nformaonall meannful reon urns ou ha he error levels are acceable. IV. COSLUSIOS An aromae connuous-me nonlnear esmaon mehod usn UKF-aded Gaussan sum reresenaons s derved. he mehod res o aromae a ror and a oseror dfs n he flern ses usn ehed Gaussan sum reresenaons. he mehod has he abl o aromae he almos eac df n a ven search area. Lookn a he smulaon resuls he mehod ves smaller error values comared h he UKF h an acceable comuaonal load and a b hher error levels han arcle Flers usn hh number of arcles for he ven ssem. he error levels and comuaonal loads of he aroach n hher dmensonal ssems and he erformance of he aroach under non- Gaussan noses are nended o be suded as a fuure ork a b c F.. AE afer 3 one Carlo runs F.. Unnormalzed aromae dfs a he end of =s for he frs samle one Carlo run: a Almos eac df; b Aromae df h UKF-aded Gaussan Sum Fler h + Gaussans; c Aromae df h UKF-aded Gaussan Sum Fler h +5 Gaussans.

6 a REFERECES [] L.L. Banasch A Comarave Sud of onlnear rackn Alorhms A dsseraon submed o he Sss Federal Insue of echnolo Zurch 99 []. Schön On Comuaonal ehods for onlnear Esmaon A dsseraon submed o he Dearmen of Elecrcal Enneern of Lnkön Unvers 3. [3] F. Daum "Eac Fne-Dmensonal onlnear Flers" IEEE ransacons on Auomac Conrol vol. AC-3 no Jul 984. [4] S. Smo On Unscened Kalman Flern for Sae Esmaon of Connuous me onlnear Ssems IEEE ransacons on Auomac Conrol vol Se. 7. [5] F. Daum onlnear Flers: Beond he Kalman Fler IEEE Aerosace and Elecronc Ssems aazne vol Au. 5. [6] S.J. Juler and J.K. Uhlmann A e Eenson of he Kalman Fler o onlnear Ssems Inroc. of AeroSense: he h In. Sm. on Aerosace/Defence Sensn Smulaon and Conrols 997. [7] A. Cubko and B. Solona Generalze Wener rocess and Kolmoorov s Equaon for dffuson nduced b non-gaussan nose source Flucaon and ose Leers vol [8] Desmond J. Hham An Alorhmc Inroducon o umercal Smulaon of Sochasc Dfferenal Equaons SIA REVIEW Vol. 43o [9] F. Daum and J. Huan Curse of Dmensonal and arcle Flers roceedns of he IEEE Aerosace Conference vol arch 3. []. Snla. Snh.D. Sco Adave Gaussan Sum Fler for onlnear Baesan Esmaon IEEE ransacons on Auomac Conrol vol Se.. []. onllo Choosn bass funcons and shae arameers for radal bass funcon mehods. SIA Underraduae Research Onlne Se b c F. 3. Unnormalzed aromae dfs a he end of =s for he second samle one Carlo run: a Almos eac df; b Aromae df h UKF-aded Gaussan Sum Fler h + Gaussans; c Aromae df h UKF-aded Gaussan Sum Fler h +5 Gaussans.

CHAPTER 10: LINEAR DISCRIMINATION

CHAPTER 10: LINEAR DISCRIMINATION CHAPER : LINEAR DISCRIMINAION Dscrmnan-based Classfcaon 3 In classfcaon h K classes (C,C,, C k ) We defned dscrmnan funcon g j (), j=,,,k hen gven an es eample, e chose (predced) s class label as C f g

More information

A New Method for Computing EM Algorithm Parameters in Speaker Identification Using Gaussian Mixture Models

A New Method for Computing EM Algorithm Parameters in Speaker Identification Using Gaussian Mixture Models 0 IACSI Hong Kong Conferences IPCSI vol. 9 (0) (0) IACSI Press, Sngaore A New ehod for Comung E Algorhm Parameers n Seaker Idenfcaon Usng Gaussan xure odels ohsen Bazyar +, Ahmad Keshavarz, and Khaoon

More information

Lecture 18: The Laplace Transform (See Sections and 14.7 in Boas)

Lecture 18: The Laplace Transform (See Sections and 14.7 in Boas) Lecure 8: The Lalace Transform (See Secons 88- and 47 n Boas) Recall ha our bg-cure goal s he analyss of he dfferenal equaon, ax bx cx F, where we emloy varous exansons for he drvng funcon F deendng on

More information

FI 3103 Quantum Physics

FI 3103 Quantum Physics /9/4 FI 33 Quanum Physcs Aleander A. Iskandar Physcs of Magnesm and Phooncs Research Grou Insu Teknolog Bandung Basc Conces n Quanum Physcs Probably and Eecaon Value Hesenberg Uncerany Prncle Wave Funcon

More information

Advanced time-series analysis (University of Lund, Economic History Department)

Advanced time-series analysis (University of Lund, Economic History Department) Advanced me-seres analss (Unvers of Lund, Economc Hsor Dearmen) 3 Jan-3 Februar and 6-3 March Lecure 4 Economerc echnues for saonar seres : Unvarae sochasc models wh Box- Jenns mehodolog, smle forecasng

More information

Foundations of State Estimation Part II

Foundations of State Estimation Part II Foundaons of Sae Esmaon Par II Tocs: Hdden Markov Models Parcle Flers Addonal readng: L.R. Rabner, A uoral on hdden Markov models," Proceedngs of he IEEE, vol. 77,. 57-86, 989. Sequenal Mone Carlo Mehods

More information

EP2200 Queuing theory and teletraffic systems. 3rd lecture Markov chains Birth-death process - Poisson process. Viktoria Fodor KTH EES

EP2200 Queuing theory and teletraffic systems. 3rd lecture Markov chains Birth-death process - Poisson process. Viktoria Fodor KTH EES EP Queung heory and eleraffc sysems 3rd lecure Marov chans Brh-deah rocess - Posson rocess Vora Fodor KTH EES Oulne for oday Marov rocesses Connuous-me Marov-chans Grah and marx reresenaon Transen and

More information

Notes on the stability of dynamic systems and the use of Eigen Values.

Notes on the stability of dynamic systems and the use of Eigen Values. Noes on he sabl of dnamc ssems and he use of Egen Values. Source: Macro II course noes, Dr. Davd Bessler s Tme Seres course noes, zarads (999) Ineremporal Macroeconomcs chaper 4 & Techncal ppend, and Hamlon

More information

Lecture VI Regression

Lecture VI Regression Lecure VI Regresson (Lnear Mehods for Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure VI: MLSC - Dr. Sehu Vjayakumar Lnear Regresson Model M

More information

NPTEL Project. Econometric Modelling. Module23: Granger Causality Test. Lecture35: Granger Causality Test. Vinod Gupta School of Management

NPTEL Project. Econometric Modelling. Module23: Granger Causality Test. Lecture35: Granger Causality Test. Vinod Gupta School of Management P age NPTEL Proec Economerc Modellng Vnod Gua School of Managemen Module23: Granger Causaly Tes Lecure35: Granger Causaly Tes Rudra P. Pradhan Vnod Gua School of Managemen Indan Insue of Technology Kharagur,

More information

Lecture 6: Learning for Control (Generalised Linear Regression)

Lecture 6: Learning for Control (Generalised Linear Regression) Lecure 6: Learnng for Conrol (Generalsed Lnear Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure 6: RLSC - Prof. Sehu Vjayakumar Lnear Regresson

More information

Outline. Energy-Efficient Target Coverage in Wireless Sensor Networks. Sensor Node. Introduction. Characteristics of WSN

Outline. Energy-Efficient Target Coverage in Wireless Sensor Networks. Sensor Node. Introduction. Characteristics of WSN Ener-Effcen Tare Coverae n Wreless Sensor Newors Presened b M Trà Tá -4-4 Inroducon Bacround Relaed Wor Our Proosal Oulne Maxmum Se Covers (MSC) Problem MSC Problem s NP-Comlee MSC Heursc Concluson Sensor

More information

J i-1 i. J i i+1. Numerical integration of the diffusion equation (I) Finite difference method. Spatial Discretization. Internal nodes.

J i-1 i. J i i+1. Numerical integration of the diffusion equation (I) Finite difference method. Spatial Discretization. Internal nodes. umercal negraon of he dffuson equaon (I) Fne dfference mehod. Spaal screaon. Inernal nodes. R L V For hermal conducon le s dscree he spaal doman no small fne spans, =,,: Balance of parcles for an nernal

More information

OP = OO' + Ut + Vn + Wb. Material We Will Cover Today. Computer Vision Lecture 3. Multi-view Geometry I. Amnon Shashua

OP = OO' + Ut + Vn + Wb. Material We Will Cover Today. Computer Vision Lecture 3. Multi-view Geometry I. Amnon Shashua Comuer Vson 27 Lecure 3 Mul-vew Geomer I Amnon Shashua Maeral We Wll Cover oa he srucure of 3D->2D rojecon mar omograh Marces A rmer on rojecve geomer of he lane Eolar Geomer an Funamenal Mar ebrew Unvers

More information

Inverse Joint Moments of Multivariate. Random Variables

Inverse Joint Moments of Multivariate. Random Variables In J Conem Mah Scences Vol 7 0 no 46 45-5 Inverse Jon Momens of Mulvarae Rom Varables M A Hussan Dearmen of Mahemacal Sascs Insue of Sascal Sudes Research ISSR Caro Unversy Egy Curren address: Kng Saud

More information

Sensor Scheduling for Multiple Parameters Estimation Under Energy Constraint

Sensor Scheduling for Multiple Parameters Estimation Under Energy Constraint Sensor Scheduln for Mulple Parameers Esmaon Under Enery Consran Y Wan, Mnyan Lu and Demoshens Tenekezs Deparmen of Elecrcal Enneern and Compuer Scence Unversy of Mchan, Ann Arbor, MI {yws,mnyan,eneke}@eecs.umch.edu

More information

THE PREDICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS

THE PREDICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS THE PREICTION OF COMPETITIVE ENVIRONMENT IN BUSINESS INTROUCTION The wo dmensonal paral dfferenal equaons of second order can be used for he smulaon of compeve envronmen n busness The arcle presens he

More information

PHYS 705: Classical Mechanics. Canonical Transformation

PHYS 705: Classical Mechanics. Canonical Transformation PHYS 705: Classcal Mechancs Canoncal Transformaon Canoncal Varables and Hamlonan Formalsm As we have seen, n he Hamlonan Formulaon of Mechancs,, are ndeenden varables n hase sace on eual foong The Hamlon

More information

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS R&RATA # Vol.) 8, March FURTHER AALYSIS OF COFIDECE ITERVALS FOR LARGE CLIET/SERVER COMPUTER ETWORKS Vyacheslav Abramov School of Mahemacal Scences, Monash Unversy, Buldng 8, Level 4, Clayon Campus, Wellngon

More information

Endogeneity. Is the term given to the situation when one or more of the regressors in the model are correlated with the error term such that

Endogeneity. Is the term given to the situation when one or more of the regressors in the model are correlated with the error term such that s row Endogeney Is he erm gven o he suaon when one or more of he regressors n he model are correlaed wh he error erm such ha E( u 0 The 3 man causes of endogeney are: Measuremen error n he rgh hand sde

More information

Existence and Uniqueness Results for Random Impulsive Integro-Differential Equation

Existence and Uniqueness Results for Random Impulsive Integro-Differential Equation Global Journal of Pure and Appled Mahemacs. ISSN 973-768 Volume 4, Number 6 (8), pp. 89-87 Research Inda Publcaons hp://www.rpublcaon.com Exsence and Unqueness Resuls for Random Impulsve Inegro-Dfferenal

More information

Pattern Classification (III) & Pattern Verification

Pattern Classification (III) & Pattern Verification Preare by Prof. Hu Jang CSE638 --4 CSE638 3. Seech & Language Processng o.5 Paern Classfcaon III & Paern Verfcaon Prof. Hu Jang Dearmen of Comuer Scence an Engneerng York Unversy Moel Parameer Esmaon Maxmum

More information

Fall 2010 Graduate Course on Dynamic Learning

Fall 2010 Graduate Course on Dynamic Learning Fall 200 Graduae Course on Dynamc Learnng Chaper 4: Parcle Flers Sepember 27, 200 Byoung-Tak Zhang School of Compuer Scence and Engneerng & Cognve Scence and Bran Scence Programs Seoul aonal Unversy hp://b.snu.ac.kr/~bzhang/

More information

Study on Multi-Target Tracking Based on Particle Filter Algorithm

Study on Multi-Target Tracking Based on Particle Filter Algorithm Research Journal of Aled Scences, Engneerng and Technology 5(2): 427-432, 213 ISSN: 24-7459; E-ISSN: 24-7467 axell Scenfc Organzaon, 213 Submed: ay 4, 212 Acceed: June 8, 212 Publshed: January 11, 213

More information

CHAPTER 10: LINEAR DISCRIMINATION

CHAPTER 10: LINEAR DISCRIMINATION HAPER : LINEAR DISRIMINAION Dscmnan-based lassfcaon 3 In classfcaon h K classes ( k ) We defned dsmnan funcon g () = K hen gven an es eample e chose (pedced) s class label as f g () as he mamum among g

More information

Normal Random Variable and its discriminant functions

Normal Random Variable and its discriminant functions Noral Rando Varable and s dscrnan funcons Oulne Noral Rando Varable Properes Dscrnan funcons Why Noral Rando Varables? Analycally racable Works well when observaon coes for a corruped snle prooype 3 The

More information

Sampling Procedure of the Sum of two Binary Markov Process Realizations

Sampling Procedure of the Sum of two Binary Markov Process Realizations Samplng Procedure of he Sum of wo Bnary Markov Process Realzaons YURY GORITSKIY Dep. of Mahemacal Modelng of Moscow Power Insue (Techncal Unversy), Moscow, RUSSIA, E-mal: gorsky@yandex.ru VLADIMIR KAZAKOV

More information

Linear Response Theory: The connection between QFT and experiments

Linear Response Theory: The connection between QFT and experiments Phys540.nb 39 3 Lnear Response Theory: The connecon beween QFT and expermens 3.1. Basc conceps and deas Q: ow do we measure he conducvy of a meal? A: we frs nroduce a weak elecrc feld E, and hen measure

More information

Chapter 6: AC Circuits

Chapter 6: AC Circuits Chaper 6: AC Crcus Chaper 6: Oulne Phasors and he AC Seady Sae AC Crcus A sable, lnear crcu operang n he seady sae wh snusodal excaon (.e., snusodal seady sae. Complee response forced response naural response.

More information

Objectives. Image R 1. Segmentation. Objects. Pixels R N. i 1 i Fall LIST 2

Objectives. Image R 1. Segmentation. Objects. Pixels R N. i 1 i Fall LIST 2 Image Segmenaon Obecves Image Pels Segmenaon R Obecs R N N R I -Fall LIS Ke Problems Feaure Sace Dsconnu and Smlar Classfer Lnear nonlnear - fuzz arallel seral -Fall LIS 3 Feaure Eracon Image Sace Feaure

More information

Part II CONTINUOUS TIME STOCHASTIC PROCESSES

Part II CONTINUOUS TIME STOCHASTIC PROCESSES Par II CONTINUOUS TIME STOCHASTIC PROCESSES 4 Chaper 4 For an advanced analyss of he properes of he Wener process, see: Revus D and Yor M: Connuous marngales and Brownan Moon Karazas I and Shreve S E:

More information

A Cell Decomposition Approach to Online Evasive Path Planning and the Video Game Ms. Pac-Man

A Cell Decomposition Approach to Online Evasive Path Planning and the Video Game Ms. Pac-Man Cell Decomoson roach o Onlne Evasve Pah Plannng and he Vdeo ame Ms. Pac-Man reg Foderaro Vram Raju Slva Ferrar Laboraory for Inellgen Sysems and Conrols LISC Dearmen of Mechancal Engneerng and Maerals

More information

A New Generalized Gronwall-Bellman Type Inequality

A New Generalized Gronwall-Bellman Type Inequality 22 Inernaonal Conference on Image, Vson and Comung (ICIVC 22) IPCSIT vol. 5 (22) (22) IACSIT Press, Sngaore DOI:.7763/IPCSIT.22.V5.46 A New Generalzed Gronwall-Bellman Tye Ineualy Qnghua Feng School of

More information

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!") i+1,q - [(!

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!) i+1,q - [(! ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL The frs hng o es n wo-way ANOVA: Is here neracon? "No neracon" means: The man effecs model would f. Ths n urn means: In he neracon plo (wh A on he horzonal

More information

3. OVERVIEW OF NUMERICAL METHODS

3. OVERVIEW OF NUMERICAL METHODS 3 OVERVIEW OF NUMERICAL METHODS 3 Inroducory remarks Ths chaper summarzes hose numercal echnques whose knowledge s ndspensable for he undersandng of he dfferen dscree elemen mehods: he Newon-Raphson-mehod,

More information

WiH Wei He

WiH Wei He Sysem Idenfcaon of onlnear Sae-Space Space Baery odels WH We He wehe@calce.umd.edu Advsor: Dr. Chaochao Chen Deparmen of echancal Engneerng Unversy of aryland, College Par 1 Unversy of aryland Bacground

More information

Hidden Markov Models with Kernel Density Estimation of Emission Probabilities and their Use in Activity Recognition

Hidden Markov Models with Kernel Density Estimation of Emission Probabilities and their Use in Activity Recognition Hdden Markov Models wh Kernel Densy Esmaon of Emsson Probables and her Use n Acvy Recognon Massmo Pccard Faculy of Informaon echnology Unversy of echnology, Sydney massmo@.us.edu.au Absrac In hs aer, we

More information

A New Approach for Solving the Unit Commitment Problem by Adaptive Particle Swarm Optimization

A New Approach for Solving the Unit Commitment Problem by Adaptive Particle Swarm Optimization A New Aroach for Solvn he Un Commmen Problem by Adave Parcle Swarm Omzaon V.S. Paala, Suden Member, IEEE, and I. Erlch, Senor Member, IEEE Absrac Ths aer resens a new aroach for formulan he un commmen

More information

Variants of Pegasos. December 11, 2009

Variants of Pegasos. December 11, 2009 Inroducon Varans of Pegasos SooWoong Ryu bshboy@sanford.edu December, 009 Youngsoo Cho yc344@sanford.edu Developng a new SVM algorhm s ongong research opc. Among many exng SVM algorhms, we wll focus on

More information

Dynamic Regressions with Variables Observed at Different Frequencies

Dynamic Regressions with Variables Observed at Different Frequencies Dynamc Regressons wh Varables Observed a Dfferen Frequences Tlak Abeysnghe and Anhony S. Tay Dearmen of Economcs Naonal Unversy of Sngaore Ken Rdge Crescen Sngaore 96 January Absrac: We consder he roblem

More information

Stochastic Maxwell Equations in Photonic Crystal Modeling and Simulations

Stochastic Maxwell Equations in Photonic Crystal Modeling and Simulations Sochasc Maxwell Equaons n Phoonc Crsal Modelng and Smulaons Hao-Mn Zhou School of Mah Georga Insue of Technolog Jon work wh: Al Adb ECE Majd Bade ECE Shu-Nee Chow Mah IPAM UCLA Aprl 14-18 2008 Parall suppored

More information

Probabilistic Lane Tracking in Difficult Road Scenarios Using Stereovision

Probabilistic Lane Tracking in Difficult Road Scenarios Using Stereovision Probablsc Lane Trackng n Dffcul Road Scenaros Usng Sereovson Radu Danescu, Sergu Nedevsch Absrac Accurae and robus lane resuls are of grea sgnfcance n any drvng asssance sysem. In order o acheve robusness

More information

Probabilistic-Fuzzy Inference Procedures for Knowledge-Based Systems

Probabilistic-Fuzzy Inference Procedures for Knowledge-Based Systems Proceedngs of he 0h WSES nernaonal Conference on MTHEMTCL and COMPUTTOL METHODS n SCECE and EGEERG (MCMESE'08 Probablsc-Fuzzy nference Procedures for Knowledge-ased Sysems WLSZEK-SZEWSK Dearmen of Conrol

More information

Density Matrix Description of NMR BCMB/CHEM 8190

Density Matrix Description of NMR BCMB/CHEM 8190 Densy Marx Descrpon of NMR BCMBCHEM 89 Operaors n Marx Noaon If we say wh one bass se, properes vary only because of changes n he coeffcens weghng each bass se funcon x = h< Ix > - hs s how we calculae

More information

Genetic Algorithm in Parameter Estimation of Nonlinear Dynamic Systems

Genetic Algorithm in Parameter Estimation of Nonlinear Dynamic Systems Genec Algorhm n Parameer Esmaon of Nonlnear Dynamc Sysems E. Paeraks manos@egnaa.ee.auh.gr V. Perds perds@vergna.eng.auh.gr Ah. ehagas kehagas@egnaa.ee.auh.gr hp://skron.conrol.ee.auh.gr/kehagas/ndex.hm

More information

HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD

HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Journal of Appled Mahemacs and Compuaonal Mechancs 3, (), 45-5 HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Sansław Kukla, Urszula Sedlecka Insue of Mahemacs,

More information

DYNAMIC ECONOMETRIC MODELS VOL. 7 NICOLAUS COPERNICUS UNIVERSITY IN TORUŃ. Jacek Kwiatkowski Nicholas Copernicus University in Toruń

DYNAMIC ECONOMETRIC MODELS VOL. 7 NICOLAUS COPERNICUS UNIVERSITY IN TORUŃ. Jacek Kwiatkowski Nicholas Copernicus University in Toruń DYAMIC ECOOMERIC MODELS VOL. 7 ICOLAUS COPERICUS UIVERSIY I ORUŃ Jacek Kwakowsk cholas Coerncus Unvers n oruń A Baesan esmaon and esng of SUR models wh alcaon o Polsh fnancal me-seres. Inroducon One of

More information

Markov Chain applications to non parametric option pricing theory

Markov Chain applications to non parametric option pricing theory IJCSS Inernaonal Journal of Comuer Scence and ewor Secury, VOL.8 o.6, June 2008 99 Marov Chan alcaons o non aramerc oon rcng heory Summary In hs aer we roose o use a Marov chan n order o rce conngen clams.

More information

FORECASTS GENERATING FOR ARCH-GARCH PROCESSES USING THE MATLAB PROCEDURES

FORECASTS GENERATING FOR ARCH-GARCH PROCESSES USING THE MATLAB PROCEDURES FORECASS GENERAING FOR ARCH-GARCH PROCESSES USING HE MALAB PROCEDURES Dušan Marček, Insiue of Comuer Science, Faculy of Philosohy and Science, he Silesian Universiy Oava he Faculy of Managemen Science

More information

On One Analytic Method of. Constructing Program Controls

On One Analytic Method of. Constructing Program Controls Appled Mahemacal Scences, Vol. 9, 05, no. 8, 409-407 HIKARI Ld, www.m-hkar.com hp://dx.do.org/0.988/ams.05.54349 On One Analyc Mehod of Consrucng Program Conrols A. N. Kvko, S. V. Chsyakov and Yu. E. Balyna

More information

Density Matrix Description of NMR BCMB/CHEM 8190

Density Matrix Description of NMR BCMB/CHEM 8190 Densy Marx Descrpon of NMR BCMBCHEM 89 Operaors n Marx Noaon Alernae approach o second order specra: ask abou x magnezaon nsead of energes and ranson probables. If we say wh one bass se, properes vary

More information

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC CH.3. COMPATIBILITY EQUATIONS Connuum Mechancs Course (MMC) - ETSECCPB - UPC Overvew Compably Condons Compably Equaons of a Poenal Vecor Feld Compably Condons for Infnesmal Srans Inegraon of he Infnesmal

More information

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005 Dynamc Team Decson Theory EECS 558 Proec Shruvandana Sharma and Davd Shuman December 0, 005 Oulne Inroducon o Team Decson Theory Decomposon of he Dynamc Team Decson Problem Equvalence of Sac and Dynamc

More information

[ ] 2. [ ]3 + (Δx i + Δx i 1 ) / 2. Δx i-1 Δx i Δx i+1. TPG4160 Reservoir Simulation 2018 Lecture note 3. page 1 of 5

[ ] 2. [ ]3 + (Δx i + Δx i 1 ) / 2. Δx i-1 Δx i Δx i+1. TPG4160 Reservoir Simulation 2018 Lecture note 3. page 1 of 5 TPG460 Reservor Smulaon 08 page of 5 DISCRETIZATIO OF THE FOW EQUATIOS As we already have seen, fne dfference appromaons of he paral dervaves appearng n he flow equaons may be obaned from Taylor seres

More information

John Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany

John Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany Herarchcal Markov Normal Mxure models wh Applcaons o Fnancal Asse Reurns Appendx: Proofs of Theorems and Condonal Poseror Dsrbuons John Geweke a and Gann Amsano b a Deparmens of Economcs and Sascs, Unversy

More information

Relative controllability of nonlinear systems with delays in control

Relative controllability of nonlinear systems with delays in control Relave conrollably o nonlnear sysems wh delays n conrol Jerzy Klamka Insue o Conrol Engneerng, Slesan Techncal Unversy, 44- Glwce, Poland. phone/ax : 48 32 37227, {jklamka}@a.polsl.glwce.pl Keywor: Conrollably.

More information

First-order piecewise-linear dynamic circuits

First-order piecewise-linear dynamic circuits Frs-order pecewse-lnear dynamc crcus. Fndng he soluon We wll sudy rs-order dynamc crcus composed o a nonlnear resse one-por, ermnaed eher by a lnear capacor or a lnear nducor (see Fg.. Nonlnear resse one-por

More information

Localization & Mapping

Localization & Mapping Auonomous Moble Robos, Chaer 5 CSE360/460-00 Inroducon o Moble Robocs Localaon & Mang Dearmen of Comuer Scence & Engneerng P.C. Rossn College of Engneerng and Aled Scence R. Segwar, I. Nourbahsh CSE360/460

More information

Machine Learning Linear Regression

Machine Learning Linear Regression Machne Learnng Lnear Regresson Lesson 3 Lnear Regresson Bascs of Regresson Leas Squares esmaon Polynomal Regresson Bass funcons Regresson model Regularzed Regresson Sascal Regresson Mamum Lkelhood (ML)

More information

Area Minimization of Power Distribution Network Using Efficient Nonlinear. Programming Techniques *

Area Minimization of Power Distribution Network Using Efficient Nonlinear. Programming Techniques * Area Mnmzaon of Power Dsrbuon Newor Usn Effcen Nonlnear Prorammn Technques * Xaoha Wu 1, Xanlon Hon 1, Yc Ca 1, C.K.Chen, Jun Gu 3 and Wayne Da 4 1 De. Of Comuer Scence and Technoloy, Tsnhua Unversy, Bejn,

More information

CHAPTER 5: MULTIVARIATE METHODS

CHAPTER 5: MULTIVARIATE METHODS CHAPER 5: MULIVARIAE MEHODS Mulvarae Daa 3 Mulple measuremens (sensors) npus/feaures/arbues: -varae N nsances/observaons/eamples Each row s an eample Each column represens a feaure X a b correspons o he

More information

グラフィカルモデルによる推論 確率伝搬法 (2) Kenji Fukumizu The Institute of Statistical Mathematics 計算推論科学概論 II (2010 年度, 後期 )

グラフィカルモデルによる推論 確率伝搬法 (2) Kenji Fukumizu The Institute of Statistical Mathematics 計算推論科学概論 II (2010 年度, 後期 ) グラフィカルモデルによる推論 確率伝搬法 Kenj Fukuzu he Insue of Sascal Maheacs 計算推論科学概論 II 年度 後期 Inference on Hdden Markov Model Inference on Hdden Markov Model Revew: HMM odel : hdden sae fne Inference Coue... for any Naïve

More information

Approximate Analytic Solution of (2+1) - Dimensional Zakharov-Kuznetsov(Zk) Equations Using Homotopy

Approximate Analytic Solution of (2+1) - Dimensional Zakharov-Kuznetsov(Zk) Equations Using Homotopy Arcle Inernaonal Journal of Modern Mahemacal Scences, 4, (): - Inernaonal Journal of Modern Mahemacal Scences Journal homepage: www.modernscenfcpress.com/journals/jmms.aspx ISSN: 66-86X Florda, USA Approxmae

More information

Performance Analysis for a Network having Standby Redundant Unit with Waiting in Repair

Performance Analysis for a Network having Standby Redundant Unit with Waiting in Repair TECHNI Inernaonal Journal of Compung Scence Communcaon Technologes VOL.5 NO. July 22 (ISSN 974-3375 erformance nalyss for a Nework havng Sby edundan Un wh ang n epar Jendra Sngh 2 abns orwal 2 Deparmen

More information

Face Detection: The Problem

Face Detection: The Problem Face Deecon and Head Trackng Yng Wu yngwu@ece.norhwesern.edu Elecrcal Engneerng & Comuer Scence Norhwesern Unversy, Evanson, IL h://www.ece.norhwesern.edu/~yngwu Face Deecon: The Problem The Goal: Idenfy

More information

GMM parameter estimation. Xiaoye Lu CMPS290c Final Project

GMM parameter estimation. Xiaoye Lu CMPS290c Final Project GMM paraeer esaon Xaoye Lu M290c Fnal rojec GMM nroducon Gaussan ure Model obnaon of several gaussan coponens Noaon: For each Gaussan dsrbuon:, s he ean and covarance ar. A GMM h ures(coponens): p ( 2π

More information

The Finite Element Method for the Analysis of Non-Linear and Dynamic Systems

The Finite Element Method for the Analysis of Non-Linear and Dynamic Systems Swss Federal Insue of Page 1 The Fne Elemen Mehod for he Analyss of Non-Lnear and Dynamc Sysems Prof. Dr. Mchael Havbro Faber Dr. Nebojsa Mojslovc Swss Federal Insue of ETH Zurch, Swzerland Mehod of Fne

More information

Pavel Azizurovich Rahman Ufa State Petroleum Technological University, Kosmonavtov St., 1, Ufa, Russian Federation

Pavel Azizurovich Rahman Ufa State Petroleum Technological University, Kosmonavtov St., 1, Ufa, Russian Federation VOL., NO. 5, MARCH 8 ISSN 89-668 ARN Journal of Engneerng and Aled Scences 6-8 Asan Research ublshng Nework ARN. All rghs reserved. www.arnjournals.com A CALCULATION METHOD FOR ESTIMATION OF THE MEAN TIME

More information

ハイブリッドモンテカルロ法に よる実現確率的ボラティリティモデルのベイズ推定

ハイブリッドモンテカルロ法に よる実現確率的ボラティリティモデルのベイズ推定 ハイブリッドモンテカルロ法に よる実現確率的ボラティリティモデルのベイズ推定 Tesuya Takas Hrosma Unversy of Economcs Oulne of resenaon 1 Inroducon Realzed volaly 3 Realzed socasc volaly 4 Bayesan nference 5 Hybrd Mone Carlo 6 Mnmum Norm negraor

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4 CS434a/54a: Paern Recognon Prof. Olga Veksler Lecure 4 Oulne Normal Random Varable Properes Dscrmnan funcons Why Normal Random Varables? Analycally racable Works well when observaon comes form a corruped

More information

January Examinations 2012

January Examinations 2012 Page of 5 EC79 January Examnaons No. of Pages: 5 No. of Quesons: 8 Subjec ECONOMICS (POSTGRADUATE) Tle of Paper EC79 QUANTITATIVE METHODS FOR BUSINESS AND FINANCE Tme Allowed Two Hours ( hours) Insrucons

More information

FTCS Solution to the Heat Equation

FTCS Solution to the Heat Equation FTCS Soluon o he Hea Equaon ME 448/548 Noes Gerald Reckenwald Porland Sae Unversy Deparmen of Mechancal Engneerng gerry@pdxedu ME 448/548: FTCS Soluon o he Hea Equaon Overvew Use he forward fne d erence

More information

Cubic Bezier Homotopy Function for Solving Exponential Equations

Cubic Bezier Homotopy Function for Solving Exponential Equations Penerb Journal of Advanced Research n Compung and Applcaons ISSN (onlne: 46-97 Vol. 4, No.. Pages -8, 6 omoopy Funcon for Solvng Eponenal Equaons S. S. Raml *,,. Mohamad Nor,a, N. S. Saharzan,b and M.

More information

Chapter 3: Signed-rank charts

Chapter 3: Signed-rank charts Chaer : gned-ran chars.. The hewhar-ye conrol char... Inroducon As menoned n Chaer, samles of fxed sze are aen a regular nervals and he long sasc s hen loed. The queson s: Whch qualy arameer should be

More information

New M-Estimator Objective Function. in Simultaneous Equations Model. (A Comparative Study)

New M-Estimator Objective Function. in Simultaneous Equations Model. (A Comparative Study) Inernaonal Mahemacal Forum, Vol. 8, 3, no., 7 - HIKARI Ld, www.m-hkar.com hp://dx.do.org/.988/mf.3.3488 New M-Esmaor Objecve Funcon n Smulaneous Equaons Model (A Comparave Sudy) Ahmed H. Youssef Professor

More information

Keywords: Hedonic regressions; hedonic indexes; consumer price indexes; superlative indexes.

Keywords: Hedonic regressions; hedonic indexes; consumer price indexes; superlative indexes. Hedonc Imuaon versus Tme Dummy Hedonc Indexes Erwn Dewer, Saeed Herav and Mck Slver December 5, 27 (wh a commenary by Jan de Haan) Dscusson Paer 7-7, Dearmen of Economcs, Unversy of Brsh Columba, 997-873

More information

Evaluation of GARCH model Adequacy in forecasting Non-linear economic time series data

Evaluation of GARCH model Adequacy in forecasting Non-linear economic time series data Journal of Comuaons & Modellng, vol.3, no., 03, -0 ISSN: 79-765 (rn), 79-8850 (onlne) Scenress Ld, 03 Evaluaon of GARCH model Adequacy n forecasng Non-lnear economc me seres daa M.O. Aknunde, P.M. Kgos

More information

Solution in semi infinite diffusion couples (error function analysis)

Solution in semi infinite diffusion couples (error function analysis) Soluon n sem nfne dffuson couples (error funcon analyss) Le us consder now he sem nfne dffuson couple of wo blocks wh concenraon of and I means ha, n a A- bnary sysem, s bondng beween wo blocks made of

More information

Density estimation. Density estimations. CS 2750 Machine Learning. Lecture 5. Milos Hauskrecht 5329 Sennott Square

Density estimation. Density estimations. CS 2750 Machine Learning. Lecture 5. Milos Hauskrecht 5329 Sennott Square Lecure 5 esy esmao Mlos Hauskrec mlos@cs..edu 539 Seo Square esy esmaos ocs: esy esmao: Mamum lkelood ML Bayesa arameer esmaes M Beroull dsrbuo. Bomal dsrbuo Mulomal dsrbuo Normal dsrbuo Eoeal famly Noaramerc

More information

MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES. Institute for Mathematical Research, Universiti Putra Malaysia, UPM Serdang, Selangor, Malaysia

MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES. Institute for Mathematical Research, Universiti Putra Malaysia, UPM Serdang, Selangor, Malaysia Malaysan Journal of Mahemacal Scences 9(2): 277-300 (2015) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES Journal homeage: h://ensemumedumy/journal A Mehod for Deermnng -Adc Orders of Facorals 1* Rafka Zulkal,

More information

Volatility Interpolation

Volatility Interpolation Volaly Inerpolaon Prelmnary Verson March 00 Jesper Andreasen and Bran Huge Danse Mares, Copenhagen wan.daddy@danseban.com brno@danseban.com Elecronc copy avalable a: hp://ssrn.com/absrac=69497 Inro Local

More information

Including the ordinary differential of distance with time as velocity makes a system of ordinary differential equations.

Including the ordinary differential of distance with time as velocity makes a system of ordinary differential equations. Soluons o Ordnary Derenal Equaons An ordnary derenal equaon has only one ndependen varable. A sysem o ordnary derenal equaons consss o several derenal equaons each wh he same ndependen varable. An eample

More information

Density estimation III.

Density estimation III. Lecure 4 esy esmao III. Mlos Hauskrec mlos@cs..edu 539 Seo Square Oule Oule: esy esmao: Mamum lkelood ML Bayesa arameer esmaes MP Beroull dsrbuo. Bomal dsrbuo Mulomal dsrbuo Normal dsrbuo Eoeal famly Eoeal

More information

PHYS 1443 Section 001 Lecture #4

PHYS 1443 Section 001 Lecture #4 PHYS 1443 Secon 001 Lecure #4 Monda, June 5, 006 Moon n Two Dmensons Moon under consan acceleraon Projecle Moon Mamum ranges and heghs Reerence Frames and relae moon Newon s Laws o Moon Force Newon s Law

More information

Department of Economics University of Warsaw Warsaw, Poland Długa Str. 44/50.

Department of Economics University of Warsaw Warsaw, Poland Długa Str. 44/50. MIGRATIOS OF HETEROGEEOUS POPULATIO OF DRIVERS ACROSS CLASSES OF A BOUS-MALUS SYSTEM BY WOJCIECH OTTO Dearmen of Economcs Unversy of Warsaw 00-24 Warsaw Poland Długa Sr. 44/50 woo@wne.uw.edu.l . ITRODUCTIO

More information

Beyond Balanced Growth : Some Further Results

Beyond Balanced Growth : Some Further Results eyond alanced Growh : Some Furher Resuls by Dens Sec and Helmu Wagner Dscusson Paer o. 49 ay 27 Dskussonsberäge der Fakulä für Wrschafswssenschaf der FernUnversä n Hagen Herausgegeben vom Dekan der Fakulä

More information

Observer Design for Nonlinear Systems using Linear Approximations

Observer Design for Nonlinear Systems using Linear Approximations Observer Desgn for Nonlnear Ssems sng Lnear Appromaons C. Navarro Hernandez, S.P. Banks and M. Aldeen Deparmen of Aomac Conrol and Ssems Engneerng, Unvers of Sheffeld, Mappn Sree, Sheffeld S 3JD. e-mal:

More information

Imperfect Information

Imperfect Information Imerfec Informaon Comlee Informaon - all layers know: Se of layers Se of sraeges for each layer Oucomes as a funcon of he sraeges Payoffs for each oucome (.e. uly funcon for each layer Incomlee Informaon

More information

A Novel Hybrid Method for Learning Bayesian Network

A Novel Hybrid Method for Learning Bayesian Network A Noel Hybrd Mehod for Learnn Bayesan Nework Wan Chun-Fen *, Lu Ku Dearmen of Mahemacs, Henan Normal Unersy, Xnxan, 4537, PR Chna * Corresondn auhor Tel: +86 1359867864; emal: wanchunfen1@16com Manuscr

More information

Lecture Slides for INTRODUCTION TO. Machine Learning. ETHEM ALPAYDIN The MIT Press,

Lecture Slides for INTRODUCTION TO. Machine Learning. ETHEM ALPAYDIN The MIT Press, Lecure Sldes for INTRDUCTIN T Machne Learnng ETHEM ALAYDIN The MIT ress, 2004 alpaydn@boun.edu.r hp://www.cmpe.boun.edu.r/~ehem/2ml CHATER 3: Hdden Marov Models Inroducon Modelng dependences n npu; no

More information

On computing differential transform of nonlinear non-autonomous functions and its applications

On computing differential transform of nonlinear non-autonomous functions and its applications On compung dfferenal ransform of nonlnear non-auonomous funcons and s applcaons Essam. R. El-Zahar, and Abdelhalm Ebad Deparmen of Mahemacs, Faculy of Scences and Humanes, Prnce Saam Bn Abdulazz Unversy,

More information

BOX-JENKINS MODEL NOTATION. The Box-Jenkins ARMA(p,q) model is denoted by the equation. pwhile the moving average (MA) part of the model is θ1at

BOX-JENKINS MODEL NOTATION. The Box-Jenkins ARMA(p,q) model is denoted by the equation. pwhile the moving average (MA) part of the model is θ1at BOX-JENKINS MODEL NOAION he Box-Jenkins ARMA(,q) model is denoed b he equaion + + L+ + a θ a L θ a 0 q q. () he auoregressive (AR) ar of he model is + L+ while he moving average (MA) ar of he model is

More information

Department of Economics University of Toronto

Department of Economics University of Toronto Deparmen of Economcs Unversy of Torono ECO408F M.A. Economercs Lecure Noes on Heeroskedascy Heeroskedascy o Ths lecure nvolves lookng a modfcaons we need o make o deal wh he regresson model when some of

More information

Preamble-Assisted Channel Estimation in OFDM-based Wireless Systems

Preamble-Assisted Channel Estimation in OFDM-based Wireless Systems reamble-asssed Channel Esmaon n OFDM-based reless Sysems Cheong-Hwan Km, Dae-Seung Ban Yong-Hwan Lee School of Elecrcal Engneerng INMC Seoul Naonal Unversy Kwanak. O. Box 34, Seoul, 5-600 Korea e-mal:

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 0 Canoncal Transformaons (Chaper 9) Wha We Dd Las Tme Hamlon s Prncple n he Hamlonan formalsm Dervaon was smple δi δ Addonal end-pon consrans pq H( q, p, ) d 0 δ q ( ) δq ( ) δ

More information

Keywords Open channel flow; Adjoint sensitivity analysis; Numerical models; Flash floods.

Keywords Open channel flow; Adjoint sensitivity analysis; Numerical models; Flash floods. Elevenh Inernaonal Waer Technoloy onference, IWT 7 Sharm El-Shekh, Ey MITIGATION OF FLAS FLOODS IN ARID REGIONS USING ADJOINT SENSITIVITY ANALYSIS ossam Elhanafy * and Graham J.M. oeland ** * PhD suden,

More information

A Unified Form for Response of 3D Generally Damped Linear Systems under Multiple Seismic Loads through Modal Analysis

A Unified Form for Response of 3D Generally Damped Linear Systems under Multiple Seismic Loads through Modal Analysis A Unfed Form for Response of 3D Generally Damped Lnear Sysems under Mulple Sesmc Loads hrouh Modal Analyss Y. hu J. Son Z. Lan 3 and G.. Lee 4 h.d. Suden Senor Research Scens 3 Research Assocae rofessor

More information

Dynamic Poverty Measures

Dynamic Poverty Measures heorecal Economcs Leers 63-69 do:436/el34 Publshed Onlne November (h://wwwscrporg/journal/el) Dynamc Povery Measures Absrac Eugene Kouass Perre Mendy Dara Seck Kern O Kymn 3 Resource Economcs Wes Vrgna

More information

A NEW TECHNIQUE FOR SOLVING THE 1-D BURGERS EQUATION

A NEW TECHNIQUE FOR SOLVING THE 1-D BURGERS EQUATION S19 A NEW TECHNIQUE FOR SOLVING THE 1-D BURGERS EQUATION by Xaojun YANG a,b, Yugu YANG a*, Carlo CATTANI c, and Mngzheng ZHU b a Sae Key Laboraory for Geomechancs and Deep Underground Engneerng, Chna Unversy

More information

The Matrix Padé Approximation in Systems of Differential Equations and Partial Differential Equations

The Matrix Padé Approximation in Systems of Differential Equations and Partial Differential Equations C Pesano-Gabno, C Gonz_Lez-Concepcon, MC Gl-Farna The Marx Padé Approxmaon n Sysems of Dfferenal Equaons and Paral Dfferenal Equaons C PESTANO-GABINO, C GONZΑLEZ-CONCEPCION, MC GIL-FARIÑA Deparmen of Appled

More information