VECTORS AND MATRICES:

Size: px
Start display at page:

Download "VECTORS AND MATRICES:"

Transcription

1 VECTORS AND MATRICES: Matrx: a rectangular array of elements Dmenson: rxc means r rows by b columns Example: A = [a ], =1,2,3; j=1,2 j In general: A = [a ], =1, r; j=1,...c j Multvarate observatons=vector: x = x1 x2... x p s a multvarate observaton. x 1,,x n s a sample of multvarate observatons. A sample can be represented by a matrx: x11,..., x1 21,..., 2 X= x x... xn1,..., x p p np Vector addton, transpose and multplcaton: T T Norm of a vector: v = (v T v) 1/2 Unt vector: v =1 Dagonal matrx: a square matrx wth all off-dagonal elements equal to 0 Identty matrx: a dagonal matrx wth all dagonal elements equal to 1 Note that A*I=I*A=A Scalar matrx: a dagonal matrx wth all dagonal elements equal to a scalar Examples:

2 3 0 0 A I I Canoncal bass {e 1,e 2,,e P } = {,,..., } Orthogonal bass, coordnate system, Gram-Schmdt: Gven a lnearly ndependent vector system convert t to an orthogonal system. 1 0 Unary vector: or 1 = Matrx multplcaton: Unary matrx: J 1 1' nn n11n 1' 1 n 1nn For addton and subtracton number of rows of the matrces must be the same and number of columns must be the same For multplcaton number of columns of the frst matrx (left multpler) must be the same as number of rows of second matrx (rght multpler) Note that A + B = B + A but A*B B*A n general

3 Determnant of a square matrx: 3 1 det det det 2det 0det Inverse Matrx: for a square matrx A s A A A -1-1 = A -1 A = I -1 such that: A s unque and has the same rank as A To have an nverse a matrx needs to be full rank (or nonsngular) To check where a matrx s of full rank we check whether ts determnant s 0 Example: 2 by 2 matrces a b 1 d b c d D c a 2 4 A D (2)(1) (4)(3) A A A D ad bc Covarance matrx: Symmetrc postve defnte Egenvalues and Egenvectors of a covarance or correlaton matrx Multvarate Normal dstrbuton has constant densty on ellpsods: s a pxp dmensonal Covarance Matrx, Ellpsods: t 1 z z c Ellpsods are characterzed also by the sem-axes drecton and length. Spectral decomposton of a covarance matrx:

4 egenvector. p vv where,=1,,p s called the -th egenvalue and v, =1,,p s called the -th 1 T Synonyms: egenvalue == sem-axs length == varance == characterstc root == latent root, egenvector==sem-axs drecton == prncpal component==characterstc vector==latent vector. In order to fnd the egenvalues and egenvectors of we use the equaton PROPERTIES v v ( Iv ) 0 det( I) 0 t () V V, where V s the matrx of egenvectors and the dagonal matrx of egenvaues of. () The egnevectors are used to descrbe the orthogonal rotaton from maxmum varance. Ths type of rotaton s often referred to as prncpal-axes rotaton. () The trace of the matrx s the sum of ts egenvlaues {λ }. (v) The determnant of the matrx s the product of the λ (v) Two egenvectors v and v j assocated wth two dstnct egenvalues λ and λ j of a symmetrc matrx are mutually orthogonal, v t v j = 0. (v) Square Root Matrx. There are several matrces that are the square root of Σ. E.g. 1/2 1/2 t V V, Cholesk factorzaton: Σ = RT R (v) Gven a set of varables X 1, X 2,...,X p, wth nonsngular covarance matrx Σ, we can always derve a set of uncorrelated varables Y 1, Y 2,...,Y p by a set of lnear transformatons correspondng to the prncpal-axes rotaton. The covarance matrx of ths new set of varables s the dagonal matrx Λ = V t ΣV (v) Gven a set of varables X 1, X 2,...,X p, wth nonsngular covarance matrx Σ x, a new set of varables Y 1, Y 2,..., Y p s defned by the transformaton Y' = X'V, where V s an orthogonal matrx. If the covarance matrx of the Y's s Σ y, then the followng relaton holds: T 1 T 1 y y y x x x In other words, the quadratc form s nvarant under rgd rotaton. Sngular value decomposton: X = UDV T where U s (n by p) orthogonal, D s dagonal, V s (p by p) orthogonal X s centered.

5 Lnear resgresson problem ( and soluton): Y n1 np p1 n1 p1 p1 (Model) Fnd b such that mn Y X Y Xb 2 2 X ' X b X ' Y (Normal Equatons) pp X b X X X Y 1 ( ' ) ' (Least Squares Estmate) Q s orthonormal f Q Q = I, Rotatons are orthonormal transformatons. cos sn R sn cos Reflexons are orthonormal transformatons: I 2 uu Sometmes people use orthogonal as orthonormal but techncally orthogonal means Q Q = dag{d} Suppose we apply an orthogonal transformaton Q to X so that X* = QX s upper trangular. Ths s QY = QX + Q or Y* =X*+, where X* s upper trangular. X* has all zeroes bellow the dagonal and bellow the pth row. * * x,..., x 11 1 p * * 0, x,..., x 22 2 p * *... X Y 1 1 Q1 X* Y* Q * * * 0,...,0, x X Y Q2 pp 2 2 0,...,0 0,...,0 Homework 2: 1. Show that b ( X ' X ) X ' Y = X Y or X b Y 1 * 1 * * * Snce X * 1 s upper trangular ths equaton can be easly solved by back substtuton. 2. Show that the transformaton HX = H p H 2 H 1 X s upper trangular by showng that each ndvdual component makes zeros bellow the dagonal of X.

6 QR decomposton: Notce also that X 1 * = Q 1 X so X= Q 1 T X 1 * Ths s called the QR decomposton of X. R Code: > A <- matrx(runf(12), 4) > A [,1] [,2] [,3] [1,] [2,] [3,] [4,] > (Q <- qr.q(qr(a))) [,1] [,2] [,3] [1,] [2,] [3,] [4,] Householder Transformatons: To construct Q we use Householder transformatons >(R <- qr.r(qr(a))) [,1] [,2] [,3] [1,] [2,] [3,] > Q%*%R [,1] [,2] [,3] [1,] [2,] [3,] [4,] > y = 1:4 > lsft(a,y,nt=f)$coef X1 X2 X > c(solve( R, t(q)%*%y)) [1] Suppose we have a nx1 vector x, we defne an orthonormal transformaton H that transforms x nto a vector x* that has zeroes at postons (+1),,n H = (I 2 u u / u 2 ) where 0... x n s 2 2 u and s x j x1 j... x n Homework Show that the transformaton HX = H p H 2 H 1 X s upper trangular by showng that each ndvdual component makes zeros bellow the dagonal of X.

7 Gram-Schmdt: Gven a lnearly ndependent vector system convert t to an orthogonal system. Let U = { u 1, u 2,,u p } be a vector system that s ndependent but not orthogonal. Step 1 Let v 1 =u 1. Step 2 Let v 2 =u 2 Proj W1 u 2 (W 1 s the subspace spanned by v 1 ) = u 2 < u 2,v 1 >/ v 1 2 v 1 Step 3 v 3 =u 3 Proj W2 u 3 (W 2 s subspace spanned by v 1 v 2 )= u 3 <u 3,v 1 >/ v 1 2 v 1 <u 3,v 2 >/ v 2 2 v 2 and so on Ths process can be also adapted to produce the QR decomposton. The problem wth Gram-Schmdt s that t becomes unstable when U s nearly sngular. Gvens Rotatons Notce the smple lnear transformaton knocks out the second component of the vector [a,b] 2 2 r a b cos sn c s a r R = R cos a / r sn cos s c b 0 sn b/ r Gvens rotatons are matrces of the form As n the smple case G(,j,) could be use to knock out the jth component of the th column of the X matrx (X ). Notce that n order to make X upper trangular we would need as many Gvens rotatons as elements bellow the dagonal are made zero. Cholesk SVD

8 Generalzaton of Least Squares GLM General Lnear models - X s generated from one or more factors by usng dummy varables. - Each factor s represented by k-1 dummes. - Column of 1 s maybe dropped - Some columns of X must be dropped n order to avod sngularty - X s lkely to be very sparse - X maybe hgh dmensonal The computatons wth GLM need to deal wth these ssues. SAS and R both do a good job wth these problems. Weghted Least Squares (WLS) Y X F V V w V w n1 np p1 n1 2 (0, ) dag{1/ } 1/ * 1/2 1/2 1/2 * * * * 2 Y V Y V X V X F I (0, ) The computatons change very lttle snce we just need to replace X and Y by ther X * and Y * Generalzed Lnear Interactve models GLIM E(Y) = ; Lnk functon: f() s lnear n X f( = X Varance functon: Var(Y) =v() Famly Lnk Mean Varance Bnomal Logt log Posson Log log Gaussan Identty Weghts are nverse to the varance: w =1/ v( ) Iteratve Reweghted Least Squares (IRLS) Algorthm ˆ X ˆ, w 1/ v( ˆ ); (1) (1) (1) (1) ˆ Use these weghts n WLS to calculate ˆ ˆ ˆ ( k ) ( k ) ( k ) ( k ) terate: X, w 1/ v( ); untl t conversges. Use these weghts n WLS to calculate ˆ (2) ( k 1)

9 Ths algorthm s very general and can be used for many problems If V ˆ s the varance functon estmate for the last teraton then ˆ ( ' ˆ X V X ) s an estmate of the covarance of ˆ. Robust regresson M- estmators y x r S( ) ( ) ( ) s s Huber's : ( z) Takng dervatvbes wth r/s to 2 z / 2 f z c 2 c z c / 2 f z c s S( ) r r 0 xj'( ) xj( ) 2 s s j z f z c Huber's : ( z) sgn( z) c f z c The dea s to wrte the system of equatons as follows: r ( ) r xj ( ) x s jr xjr w s r =0 In matrx terms ths gves X Wr=0 or X WY=X W The soluton s ˆ =(X WX) -1 X WY But of course the matrx W does depend on so we apply here the same teratve reweghted least squares algorthm of the prevous secton. At convergence we obtan our estmate ˆ of Nonlnear Methods. Optmzaton methods. Maxmum Lkelhood estmaton. L x () = P( X ) l x () = ln(p( X )) ˆ maxmzes L ( ), or l ( ). The nformaton matrx s defned as x I( ) [ E l ( )] and asymptotcally converges to the nverse covarance matrx of ˆ X asy var( ˆ ) I( ) where s the true value of the parameter x

10 Optmzaton methods Newton-Raphson: Solve f(x) = 0 starng an ntal value close to the soluton. x +1 = x f(x ) / f (x ) Newton-Raphson usually converges very fast Multvarate verson f(x) = 0 0 = f(x k ) + f (x k ) (s - x k ) + x k+1 ~ x k - [ f (x k ) ] -1 f(x k ) px1 px1 pxp px1 Do a search along the drecton of d and fnd a mnmum. Steepest descent: d S = -F (x) Newton Steps: d N = - [F (x)] -1 F (x) Levenberg-Marquardt : d LM = - [F (x)+ I] -1 F (x) for > 0 Smplex Method (Nelder Mead) Use functons optmze and optm n R. Example: f <- functon(x) felse(x > -1, felse(x < 4, exp(-1/abs(x - 1)), 10), 10) fp <- functon(x) { prnt(x); f(x) } Example: plot(f, -2,5, ylm = 0:1, col = 2) optmze(fp, c(-4, 20))# doesn't see the mnmum optmze(fp, c(-7, 20))# ok fr <- functon(x) { ## Rosenbrock Banana functon x1 <- x[1] x2 <- x[2] 100 * (x2 - x1 * x1)^2 + (1 - x1)^2 } grr <- functon(x) { ## Gradent of 'fr' x1 <- x[1] x2 <- x[2] c(-400 * x1 * (x2 - x1 * x1) - 2 * (1 - x1), 200 * (x2 - x1 * x1)) }

11 optm(c(-1.2,1), fr) optm(c(-1.2,1), fr, grr, method = "BFGS") optm(c(-1.2,1), fr, NULL, method = "BFGS", hessan = TRUE) optm(c(-1.2,1), fr, grr, method = "CG") optm(c(-1.2,1), fr, grr, method = "CG", control=lst(type=2)) optm(c(-1.2,1), fr, grr, method = "L-BFGS-B") Non Lnear Least squares: Remember that when the error term s addtve and d normal then NLS s the same as maxmum lkelhood. These are the steps of the Gauss-Newton algorthm. t 2 y g( x, ) y g ( ) g ( ) ( ) g( ) F(0, ) y g ( ) ( ) g( ) Y X ( ) t t Gauss-Newton Method: ( ) ( X X ) X Y [ ( g)( g) ] g( y g ) G F 0 t 1 t t 1 1 Iterate ths equalty untl convergence. To do that at each teraton replace wth the new. 0 Example: Emax model Emax D g( D) EY D E, 0 ED D 50 where E 0 s the response Y at baselne (absence of dose), E max s the asymptotc maxmum dose effect and ED 50 s the dose whch produces 50% of the maxmal effect. A generalzaton of equaton above s the 4-parameter EMAX model for E g( D) E Y D E ED max 0 where s the 4 th parameter whch s sometmes called the Hll parameter (Holford and Shener). The Hll parameter affects the shape of the curve and s n some cases very dffcult to estmate. In Fgure 1 we show curves for a range of Hll parameter values. 50 D D, In fttng models above the error dstrbuton s assumed to be d normal wth zero mean and equal varance. Ths assumpton s not n any way a restrcton of the methods that we dscuss here but a convenent assumpton that smplfes the ssues that we address. Our model becomes: Y = g (D) + where =(E 0,ED 50,EMAX, and ~d N(0, ).

12 Fgure: Curve Shapes for Dfferent Values of Hll Parameter. Fgure above llustrates the rch varety of shapes generated by changng the Hll parameter. For the curves have concave downward shapes whereas for the curves have a sgmodal shape. For very small the curve represents a flat response for any dose greater than zero, whereas for very large the curve approaches a step functon at ED Maxmum Lkelhood Estmaton for the EMAX model.

13 Under the Emax model the MLE estmator of s the least squares estmator ˆ. One way to calculate ˆ s by non-lnear least squares mnmzaton (NLS). Gven the data {(D, y ), =1,,N} the NLS estmator mnmzes (n ) the followng quantty: SSE()= N 1 ( y g ( D )) The algorthm to mnmze ths quantty s a specal case of Newton-Raphson specalzed for mnmzng sums of squares. It s mplemented n many standard software packages such as R, S-plus and others. The NLS mplementaton of the MLE s s convenent because t allows the computaton of confdence ntervals for the model parameters and p-values for testng the sgnfcance of the parameters. 2 Example. Dose response estmaton n a clncal tral. In order to llustrate the ssues that we address n ths paper we use data from an unpublshed phase II clncal tral. The study had 5 doses (0, 5, 25, 50 and 100 mg) and correspondng group szes (n=78, 81, 81, 81, 77). Unlke most dose-response trals ths study has about 80 patents per arm and a hgh sgnal to nose rato. The Fgure below plots the response aganst the doses wth 90% confdence ntervals around the mean. The graph shows that t s a potentally good canddate for the EMAX model. We frst ft 3-Paramater EMAX model to the data. The maxmum lkelhood estmators (MLEs) converge quckly and have estmates of Ê max = 15.13, Ê 0 = 2.47, and ED 50 = Fgure: Clncal tral example wth 5 doses. Red stars represent the response at a gven dose and 90% CI. Sold lne s the 3-parameter ft. Dashed lne refers to the 4-parameter ft (usng the estmates after 100 teratons.)

14 Change n EDD Parameter EMAX 4 Parameter EMAX Bayesan Approach Power Law Dose (mg) One concern about the ft n the fgure s that the 3 parameter EMAX ft lacks the curvature necessary to ft well wth the observed relatonshp and the resduals show an nverted U pattern. To address the above lack of ft we next ft a 4-parameter EMAX model, but n ths case we encounter non-convergence of the MLEs. Estmates of ED 50, and E max do not converge. The Hll parameter,, trends toward a value less than 1. If we trace the values of the MLEs after a few hundred teratons of the NLS algorthm we observe Ê max = 2787, Ê 0 = 2.0, ˆ=0.52 and ED 50 = wth large standard errors. The sequence of ED 50 s and E max s are monotoncally ncreasng at a faster speed at each teraton untl the computaton becomes unstable (algorthm fals). At the same tme the mean squared error (MSE) decreases very slowly towards an asymptote. Ths suggests that the lkelhood, although bounded, converges to the maxmum as both ED 50 and E max go to. R CODE: m = nls(ch8~e0+(emax*dose)/(dose+ed50),start= c(ed50=25,e0=5,emax=20), control=nls.control(maxter=100),trace=t,na.acton=na.omt,data=edd) f1 = ftted(m) sg = summary(m)$sgma

15 Stochastc Approxmaton. Solve an equaton f(x) = 0 but nstead of observng f(x) we observe g(x) = f(x) + error The algorthm s due to Robns Monrow(1955) x +1 = x +1 - c y / Example: Logstc regresson E ( ˆ ) = ˆ or E ( ˆ ) - ˆ = ˆ 1 ˆ ˆ ˆ 1 ( E ˆ ( ) - 0) Homework: 1. Gven the followng data x= x = c( -10,(-3):5,10) ; y = c(0,0,1,0,0,1,0,1,1,1,1) calculate the stochastc approxmaton estmator of beta. Expected value functon ef =functon(b,z=x,ns=100) { n = length(x); px = 1/(1+ exp(-b*z) ) y = array( (runf(n*ns) <=px)*1, c(n,ns) ) for( n 1:ns) n[] = glm(y[,]~z-1,famly=bnomal)$coef medan(n) } Random number generaton Samplng from known CDF's. Smulatons.

Linear Approximation with Regularization and Moving Least Squares

Linear Approximation with Regularization and Moving Least Squares Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...

More information

Composite Hypotheses testing

Composite Hypotheses testing Composte ypotheses testng In many hypothess testng problems there are many possble dstrbutons that can occur under each of the hypotheses. The output of the source s a set of parameters (ponts n a parameter

More information

Singular Value Decomposition: Theory and Applications

Singular Value Decomposition: Theory and Applications Sngular Value Decomposton: Theory and Applcatons Danel Khashab Sprng 2015 Last Update: March 2, 2015 1 Introducton A = UDV where columns of U and V are orthonormal and matrx D s dagonal wth postve real

More information

APPENDIX A Some Linear Algebra

APPENDIX A Some Linear Algebra APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,

More information

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 16

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2018 LECTURE 16 STAT 39: MATHEMATICAL COMPUTATIONS I FALL 218 LECTURE 16 1 why teratve methods f we have a lnear system Ax = b where A s very, very large but s ether sparse or structured (eg, banded, Toepltz, banded plus

More information

Some Comments on Accelerating Convergence of Iterative Sequences Using Direct Inversion of the Iterative Subspace (DIIS)

Some Comments on Accelerating Convergence of Iterative Sequences Using Direct Inversion of the Iterative Subspace (DIIS) Some Comments on Acceleratng Convergence of Iteratve Sequences Usng Drect Inverson of the Iteratve Subspace (DIIS) C. Davd Sherrll School of Chemstry and Bochemstry Georga Insttute of Technology May 1998

More information

Inner Product. Euclidean Space. Orthonormal Basis. Orthogonal

Inner Product. Euclidean Space. Orthonormal Basis. Orthogonal Inner Product Defnton 1 () A Eucldean space s a fnte-dmensonal vector space over the reals R, wth an nner product,. Defnton 2 (Inner Product) An nner product, on a real vector space X s a symmetrc, blnear,

More information

IV. Performance Optimization

IV. Performance Optimization IV. Performance Optmzaton A. Steepest descent algorthm defnton how to set up bounds on learnng rate mnmzaton n a lne (varyng learnng rate) momentum learnng examples B. Newton s method defnton Gauss-Newton

More information

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could

More information

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U)

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U) Econ 413 Exam 13 H ANSWERS Settet er nndelt 9 deloppgaver, A,B,C, som alle anbefales å telle lkt for å gøre det ltt lettere å stå. Svar er gtt . Unfortunately, there s a prntng error n the hnt of

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

MEM 255 Introduction to Control Systems Review: Basics of Linear Algebra

MEM 255 Introduction to Control Systems Review: Basics of Linear Algebra MEM 255 Introducton to Control Systems Revew: Bascs of Lnear Algebra Harry G. Kwatny Department of Mechancal Engneerng & Mechancs Drexel Unversty Outlne Vectors Matrces MATLAB Advanced Topcs Vectors A

More information

Feb 14: Spatial analysis of data fields

Feb 14: Spatial analysis of data fields Feb 4: Spatal analyss of data felds Mappng rregularly sampled data onto a regular grd Many analyss technques for geophyscal data requre the data be located at regular ntervals n space and/or tme. hs s

More information

Salmon: Lectures on partial differential equations. Consider the general linear, second-order PDE in the form. ,x 2

Salmon: Lectures on partial differential equations. Consider the general linear, second-order PDE in the form. ,x 2 Salmon: Lectures on partal dfferental equatons 5. Classfcaton of second-order equatons There are general methods for classfyng hgher-order partal dfferental equatons. One s very general (applyng even to

More information

Maximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models

Maximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models ECO 452 -- OE 4: Probt and Logt Models ECO 452 -- OE 4 Maxmum Lkelhood Estmaton of Bnary Dependent Varables Models: Probt and Logt hs note demonstrates how to formulate bnary dependent varables models

More information

Chapter 7 Generalized and Weighted Least Squares Estimation. In this method, the deviation between the observed and expected values of

Chapter 7 Generalized and Weighted Least Squares Estimation. In this method, the deviation between the observed and expected values of Chapter 7 Generalzed and Weghted Least Squares Estmaton The usual lnear regresson model assumes that all the random error components are dentcally and ndependently dstrbuted wth constant varance. When

More information

= = = (a) Use the MATLAB command rref to solve the system. (b) Let A be the coefficient matrix and B be the right-hand side of the system.

= = = (a) Use the MATLAB command rref to solve the system. (b) Let A be the coefficient matrix and B be the right-hand side of the system. Chapter Matlab Exercses Chapter Matlab Exercses. Consder the lnear system of Example n Secton.. x x x y z y y z (a) Use the MATLAB command rref to solve the system. (b) Let A be the coeffcent matrx and

More information

MLE and Bayesian Estimation. Jie Tang Department of Computer Science & Technology Tsinghua University 2012

MLE and Bayesian Estimation. Jie Tang Department of Computer Science & Technology Tsinghua University 2012 MLE and Bayesan Estmaton Je Tang Department of Computer Scence & Technology Tsnghua Unversty 01 1 Lnear Regresson? As the frst step, we need to decde how we re gong to represent the functon f. One example:

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 31 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 6. Rdge regresson The OLSE s the best lnear unbased

More information

CIS526: Machine Learning Lecture 3 (Sept 16, 2003) Linear Regression. Preparation help: Xiaoying Huang. x 1 θ 1 output... θ M x M

CIS526: Machine Learning Lecture 3 (Sept 16, 2003) Linear Regression. Preparation help: Xiaoying Huang. x 1 θ 1 output... θ M x M CIS56: achne Learnng Lecture 3 (Sept 6, 003) Preparaton help: Xaoyng Huang Lnear Regresson Lnear regresson can be represented by a functonal form: f(; θ) = θ 0 0 +θ + + θ = θ = 0 ote: 0 s a dummy attrbute

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017 U.C. Berkeley CS94: Beyond Worst-Case Analyss Handout 4s Luca Trevsan September 5, 07 Summary of Lecture 4 In whch we ntroduce semdefnte programmng and apply t to Max Cut. Semdefnte Programmng Recall that

More information

Review: Fit a line to N data points

Review: Fit a line to N data points Revew: Ft a lne to data ponts Correlated parameters: L y = a x + b Orthogonal parameters: J y = a (x ˆ x + b For ntercept b, set a=0 and fnd b by optmal average: ˆ b = y, Var[ b ˆ ] = For slope a, set

More information

Vector Norms. Chapter 7 Iterative Techniques in Matrix Algebra. Cauchy-Bunyakovsky-Schwarz Inequality for Sums. Distances. Convergence.

Vector Norms. Chapter 7 Iterative Techniques in Matrix Algebra. Cauchy-Bunyakovsky-Schwarz Inequality for Sums. Distances. Convergence. Vector Norms Chapter 7 Iteratve Technques n Matrx Algebra Per-Olof Persson persson@berkeley.edu Department of Mathematcs Unversty of Calforna, Berkeley Math 128B Numercal Analyss Defnton A vector norm

More information

Inexact Newton Methods for Inverse Eigenvalue Problems

Inexact Newton Methods for Inverse Eigenvalue Problems Inexact Newton Methods for Inverse Egenvalue Problems Zheng-jan Ba Abstract In ths paper, we survey some of the latest development n usng nexact Newton-lke methods for solvng nverse egenvalue problems.

More information

Chapter 14: Logit and Probit Models for Categorical Response Variables

Chapter 14: Logit and Probit Models for Categorical Response Variables Chapter 4: Logt and Probt Models for Categorcal Response Varables Sect 4. Models for Dchotomous Data We wll dscuss only ths secton of Chap 4, whch s manly about Logstc Regresson, a specal case of the famly

More information

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 30 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 2 Remedes for multcollnearty Varous technques have

More information

Solutions to exam in SF1811 Optimization, Jan 14, 2015

Solutions to exam in SF1811 Optimization, Jan 14, 2015 Solutons to exam n SF8 Optmzaton, Jan 4, 25 3 3 O------O -4 \ / \ / The network: \/ where all lnks go from left to rght. /\ / \ / \ 6 O------O -5 2 4.(a) Let x = ( x 3, x 4, x 23, x 24 ) T, where the varable

More information

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur Analyss of Varance and Desgn of Experment-I MODULE VII LECTURE - 3 ANALYSIS OF COVARIANCE Dr Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur Any scentfc experment s performed

More information

Radar Trackers. Study Guide. All chapters, problems, examples and page numbers refer to Applied Optimal Estimation, A. Gelb, Ed.

Radar Trackers. Study Guide. All chapters, problems, examples and page numbers refer to Applied Optimal Estimation, A. Gelb, Ed. Radar rackers Study Gude All chapters, problems, examples and page numbers refer to Appled Optmal Estmaton, A. Gelb, Ed. Chapter Example.0- Problem Statement wo sensors Each has a sngle nose measurement

More information

Feature Selection: Part 1

Feature Selection: Part 1 CSE 546: Machne Learnng Lecture 5 Feature Selecton: Part 1 Instructor: Sham Kakade 1 Regresson n the hgh dmensonal settng How do we learn when the number of features d s greater than the sample sze n?

More information

Optimization. September 4, 2018

Optimization. September 4, 2018 Optmzaton September 4, 2018 Optmzaton problem 1/34 An optmzaton problem s the problem of fndng the best soluton for an objectve functon. Optmzaton method plays an mportant role n statstcs, for example,

More information

However, since P is a symmetric idempotent matrix, of P are either 0 or 1 [Eigen-values

However, since P is a symmetric idempotent matrix, of P are either 0 or 1 [Eigen-values Fall 007 Soluton to Mdterm Examnaton STAT 7 Dr. Goel. [0 ponts] For the general lnear model = X + ε, wth uncorrelated errors havng mean zero and varance σ, suppose that the desgn matrx X s not necessarly

More information

Predictive Analytics : QM901.1x Prof U Dinesh Kumar, IIMB. All Rights Reserved, Indian Institute of Management Bangalore

Predictive Analytics : QM901.1x Prof U Dinesh Kumar, IIMB. All Rights Reserved, Indian Institute of Management Bangalore Sesson Outlne Introducton to classfcaton problems and dscrete choce models. Introducton to Logstcs Regresson. Logstc functon and Logt functon. Maxmum Lkelhood Estmator (MLE) for estmaton of LR parameters.

More information

Tracking with Kalman Filter

Tracking with Kalman Filter Trackng wth Kalman Flter Scott T. Acton Vrgna Image and Vdeo Analyss (VIVA), Charles L. Brown Department of Electrcal and Computer Engneerng Department of Bomedcal Engneerng Unversty of Vrgna, Charlottesvlle,

More information

Fall 2012 Analysis of Experimental Measurements B. Eisenstein/rev. S. Errede

Fall 2012 Analysis of Experimental Measurements B. Eisenstein/rev. S. Errede Fall 0 Analyss of Expermental easurements B. Esensten/rev. S. Errede We now reformulate the lnear Least Squares ethod n more general terms, sutable for (eventually extendng to the non-lnear case, and also

More information

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur

Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur Analyss of Varance and Desgn of Exerments-I MODULE III LECTURE - 2 EXPERIMENTAL DESIGN MODELS Dr. Shalabh Deartment of Mathematcs and Statstcs Indan Insttute of Technology Kanur 2 We consder the models

More information

STAT 405 BIOSTATISTICS (Fall 2016) Handout 15 Introduction to Logistic Regression

STAT 405 BIOSTATISTICS (Fall 2016) Handout 15 Introduction to Logistic Regression STAT 45 BIOSTATISTICS (Fall 26) Handout 5 Introducton to Logstc Regresson Ths handout covers materal found n Secton 3.7 of your text. You may also want to revew regresson technques n Chapter. In ths handout,

More information

Negative Binomial Regression

Negative Binomial Regression STATGRAPHICS Rev. 9/16/2013 Negatve Bnomal Regresson Summary... 1 Data Input... 3 Statstcal Model... 3 Analyss Summary... 4 Analyss Optons... 7 Plot of Ftted Model... 8 Observed Versus Predcted... 10 Predctons...

More information

Maximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models

Maximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models ECO 452 -- OE 4: Probt and Logt Models ECO 452 -- OE 4 Mamum Lkelhood Estmaton of Bnary Dependent Varables Models: Probt and Logt hs note demonstrates how to formulate bnary dependent varables models for

More information

Chat eld, C. and A.J.Collins, Introduction to multivariate analysis. Chapman & Hall, 1980

Chat eld, C. and A.J.Collins, Introduction to multivariate analysis. Chapman & Hall, 1980 MT07: Multvarate Statstcal Methods Mke Tso: emal mke.tso@manchester.ac.uk Webpage for notes: http://www.maths.manchester.ac.uk/~mkt/new_teachng.htm. Introducton to multvarate data. Books Chat eld, C. and

More information

Generalized Linear Methods

Generalized Linear Methods Generalzed Lnear Methods 1 Introducton In the Ensemble Methods the general dea s that usng a combnaton of several weak learner one could make a better learner. More formally, assume that we have a set

More information

Optimization. August 30, 2016

Optimization. August 30, 2016 Optmzaton August 30, 2016 Optmzaton problem 1/31 An optmzaton problem s the problem of fndng the best soluton for an objectve functon. Optmzaton method plays an mportant role n statstcs, for example, to

More information

2.3 Nilpotent endomorphisms

2.3 Nilpotent endomorphisms s a block dagonal matrx, wth A Mat dm U (C) In fact, we can assume that B = B 1 B k, wth B an ordered bass of U, and that A = [f U ] B, where f U : U U s the restrcton of f to U 40 23 Nlpotent endomorphsms

More information

The Prncpal Component Transform The Prncpal Component Transform s also called Karhunen-Loeve Transform (KLT, Hotellng Transform, oregenvector Transfor

The Prncpal Component Transform The Prncpal Component Transform s also called Karhunen-Loeve Transform (KLT, Hotellng Transform, oregenvector Transfor Prncpal Component Transform Multvarate Random Sgnals A real tme sgnal x(t can be consdered as a random process and ts samples x m (m =0; ;N, 1 a random vector: The mean vector of X s X =[x0; ;x N,1] T

More information

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0 MODULE 2 Topcs: Lnear ndependence, bass and dmenson We have seen that f n a set of vectors one vector s a lnear combnaton of the remanng vectors n the set then the span of the set s unchanged f that vector

More information

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number

More information

5 The Rational Canonical Form

5 The Rational Canonical Form 5 The Ratonal Canoncal Form Here p s a monc rreducble factor of the mnmum polynomal m T and s not necessarly of degree one Let F p denote the feld constructed earler n the course, consstng of all matrces

More information

Estimating the Fundamental Matrix by Transforming Image Points in Projective Space 1

Estimating the Fundamental Matrix by Transforming Image Points in Projective Space 1 Estmatng the Fundamental Matrx by Transformng Image Ponts n Projectve Space 1 Zhengyou Zhang and Charles Loop Mcrosoft Research, One Mcrosoft Way, Redmond, WA 98052, USA E-mal: fzhang,cloopg@mcrosoft.com

More information

10-701/ Machine Learning, Fall 2005 Homework 3

10-701/ Machine Learning, Fall 2005 Homework 3 10-701/15-781 Machne Learnng, Fall 2005 Homework 3 Out: 10/20/05 Due: begnnng of the class 11/01/05 Instructons Contact questons-10701@autonlaborg for queston Problem 1 Regresson and Cross-valdaton [40

More information

LECTURE 9 CANONICAL CORRELATION ANALYSIS

LECTURE 9 CANONICAL CORRELATION ANALYSIS LECURE 9 CANONICAL CORRELAION ANALYSIS Introducton he concept of canoncal correlaton arses when we want to quantfy the assocatons between two sets of varables. For example, suppose that the frst set of

More information

The conjugate prior to a Bernoulli is. A) Bernoulli B) Gaussian C) Beta D) none of the above

The conjugate prior to a Bernoulli is. A) Bernoulli B) Gaussian C) Beta D) none of the above The conjugate pror to a Bernoull s A) Bernoull B) Gaussan C) Beta D) none of the above The conjugate pror to a Gaussan s A) Bernoull B) Gaussan C) Beta D) none of the above MAP estmates A) argmax θ p(θ

More information

Chapter 11: Simple Linear Regression and Correlation

Chapter 11: Simple Linear Regression and Correlation Chapter 11: Smple Lnear Regresson and Correlaton 11-1 Emprcal Models 11-2 Smple Lnear Regresson 11-3 Propertes of the Least Squares Estmators 11-4 Hypothess Test n Smple Lnear Regresson 11-4.1 Use of t-tests

More information

Quantum Mechanics I - Session 4

Quantum Mechanics I - Session 4 Quantum Mechancs I - Sesson 4 Aprl 3, 05 Contents Operators Change of Bass 4 3 Egenvectors and Egenvalues 5 3. Denton....................................... 5 3. Rotaton n D....................................

More information

Chapter 2 - The Simple Linear Regression Model S =0. e i is a random error. S β2 β. This is a minimization problem. Solution is a calculus exercise.

Chapter 2 - The Simple Linear Regression Model S =0. e i is a random error. S β2 β. This is a minimization problem. Solution is a calculus exercise. Chapter - The Smple Lnear Regresson Model The lnear regresson equaton s: where y + = β + β e for =,..., y and are observable varables e s a random error How can an estmaton rule be constructed for the

More information

[ ] λ λ λ. Multicollinearity. multicollinearity Ragnar Frisch (1934) perfect exact. collinearity. multicollinearity. exact

[ ] λ λ λ. Multicollinearity. multicollinearity Ragnar Frisch (1934) perfect exact. collinearity. multicollinearity. exact Multcollnearty multcollnearty Ragnar Frsch (934 perfect exact collnearty multcollnearty K exact λ λ λ K K x+ x+ + x 0 0.. λ, λ, λk 0 0.. x perfect ntercorrelated λ λ λ x+ x+ + KxK + v 0 0.. v 3 y β + β

More information

Lecture 12: Classification

Lecture 12: Classification Lecture : Classfcaton g Dscrmnant functons g The optmal Bayes classfer g Quadratc classfers g Eucldean and Mahalanobs metrcs g K Nearest Neghbor Classfers Intellgent Sensor Systems Rcardo Guterrez-Osuna

More information

Generative classification models

Generative classification models CS 675 Intro to Machne Learnng Lecture Generatve classfcaton models Mlos Hauskrecht mlos@cs.ptt.edu 539 Sennott Square Data: D { d, d,.., dn} d, Classfcaton represents a dscrete class value Goal: learn

More information

e i is a random error

e i is a random error Chapter - The Smple Lnear Regresson Model The lnear regresson equaton s: where + β + β e for,..., and are observable varables e s a random error How can an estmaton rule be constructed for the unknown

More information

EEE 241: Linear Systems

EEE 241: Linear Systems EEE : Lnear Systems Summary #: Backpropagaton BACKPROPAGATION The perceptron rule as well as the Wdrow Hoff learnng were desgned to tran sngle layer networks. They suffer from the same dsadvantage: they

More information

Limited Dependent Variables

Limited Dependent Variables Lmted Dependent Varables. What f the left-hand sde varable s not a contnuous thng spread from mnus nfnty to plus nfnty? That s, gven a model = f (, β, ε, where a. s bounded below at zero, such as wages

More information

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS

8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 493 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces you have studed thus far n the text are real vector spaces because the scalars

More information

Topic 5: Non-Linear Regression

Topic 5: Non-Linear Regression Topc 5: Non-Lnear Regresson The models we ve worked wth so far have been lnear n the parameters. They ve been of the form: y = Xβ + ε Many models based on economc theory are actually non-lnear n the parameters.

More information

Parametric fractional imputation for missing data analysis. Jae Kwang Kim Survey Working Group Seminar March 29, 2010

Parametric fractional imputation for missing data analysis. Jae Kwang Kim Survey Working Group Seminar March 29, 2010 Parametrc fractonal mputaton for mssng data analyss Jae Kwang Km Survey Workng Group Semnar March 29, 2010 1 Outlne Introducton Proposed method Fractonal mputaton Approxmaton Varance estmaton Multple mputaton

More information

Chapter Newton s Method

Chapter Newton s Method Chapter 9. Newton s Method After readng ths chapter, you should be able to:. Understand how Newton s method s dfferent from the Golden Secton Search method. Understand how Newton s method works 3. Solve

More information

WORKSHOP ON THE ANALYSIS OF NEURAL DATA 2001 MARINE BIOLOGICAL LABORATORY WOODS HOLE, MASSACHUSETTS

WORKSHOP ON THE ANALYSIS OF NEURAL DATA 2001 MARINE BIOLOGICAL LABORATORY WOODS HOLE, MASSACHUSETTS Emery. Brown, M.D., Ph.D. WORKSHOP O THE AALYSIS OF EURAL DATA 00 MARIE BIOLOGICAL LABORATORY WOODS HOLE, MASSACHUSETTS A REVIEW OF STATISTICS PART 3: REGRESSIO AALYSIS AD THE GEERALIZED LIEAR MODEL EMERY.

More information

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve

More information

Lecture 4: September 12

Lecture 4: September 12 36-755: Advanced Statstcal Theory Fall 016 Lecture 4: September 1 Lecturer: Alessandro Rnaldo Scrbe: Xao Hu Ta Note: LaTeX template courtesy of UC Berkeley EECS dept. Dsclamer: These notes have not been

More information

MEM633 Lectures 7&8. Chapter 4. Descriptions of MIMO Systems 4-1 Direct Realizations. (i) x u. y x

MEM633 Lectures 7&8. Chapter 4. Descriptions of MIMO Systems 4-1 Direct Realizations. (i) x u. y x MEM633 Lectures 7&8 Chapter 4 Descrptons of MIMO Systems 4- Drect ealzatons y() s s su() s y () s u () s ( s)( s) s y() s u (), s y() s u() s s s y() s u(), s y() s u() s ( s)( s) s () ( s ) y ( s) u (

More information

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity Week3, Chapter 4 Moton n Two Dmensons Lecture Quz A partcle confned to moton along the x axs moves wth constant acceleraton from x =.0 m to x = 8.0 m durng a 1-s tme nterval. The velocty of the partcle

More information

Probability Theory (revisited)

Probability Theory (revisited) Probablty Theory (revsted) Summary Probablty v.s. plausblty Random varables Smulaton of Random Experments Challenge The alarm of a shop rang. Soon afterwards, a man was seen runnng n the street, persecuted

More information

Logistic regression models 1/12

Logistic regression models 1/12 Logstc regresson models 1/12 2/12 Example 1: dogs look lke ther owners? Some people beleve that dogs look lke ther owners. Is ths true? To test the above hypothess, The New York Tmes conducted a quz onlne.

More information

CSCE 790S Background Results

CSCE 790S Background Results CSCE 790S Background Results Stephen A. Fenner September 8, 011 Abstract These results are background to the course CSCE 790S/CSCE 790B, Quantum Computaton and Informaton (Sprng 007 and Fall 011). Each

More information

The exam is closed book, closed notes except your one-page cheat sheet.

The exam is closed book, closed notes except your one-page cheat sheet. CS 89 Fall 206 Introducton to Machne Learnng Fnal Do not open the exam before you are nstructed to do so The exam s closed book, closed notes except your one-page cheat sheet Usage of electronc devces

More information

where I = (n x n) diagonal identity matrix with diagonal elements = 1 and off-diagonal elements = 0; and σ 2 e = variance of (Y X).

where I = (n x n) diagonal identity matrix with diagonal elements = 1 and off-diagonal elements = 0; and σ 2 e = variance of (Y X). 11.4.1 Estmaton of Multple Regresson Coeffcents In multple lnear regresson, we essentally solve n equatons for the p unnown parameters. hus n must e equal to or greater than p and n practce n should e

More information

Homework Notes Week 7

Homework Notes Week 7 Homework Notes Week 7 Math 4 Sprng 4 #4 (a Complete the proof n example 5 that s an nner product (the Frobenus nner product on M n n (F In the example propertes (a and (d have already been verfed so we

More information

Logistic Regression Maximum Likelihood Estimation

Logistic Regression Maximum Likelihood Estimation Harvard-MIT Dvson of Health Scences and Technology HST.951J: Medcal Decson Support, Fall 2005 Instructors: Professor Lucla Ohno-Machado and Professor Staal Vnterbo 6.873/HST.951 Medcal Decson Support Fall

More information

Dynamic Systems on Graphs

Dynamic Systems on Graphs Prepared by F.L. Lews Updated: Saturday, February 06, 200 Dynamc Systems on Graphs Control Graphs and Consensus A network s a set of nodes that collaborates to acheve what each cannot acheve alone. A network,

More information

Statistical pattern recognition

Statistical pattern recognition Statstcal pattern recognton Bayes theorem Problem: decdng f a patent has a partcular condton based on a partcular test However, the test s mperfect Someone wth the condton may go undetected (false negatve

More information

Time-Varying Systems and Computations Lecture 6

Time-Varying Systems and Computations Lecture 6 Tme-Varyng Systems and Computatons Lecture 6 Klaus Depold 14. Januar 2014 The Kalman Flter The Kalman estmaton flter attempts to estmate the actual state of an unknown dscrete dynamcal system, gven nosy

More information

Classification as a Regression Problem

Classification as a Regression Problem Target varable y C C, C,, ; Classfcaton as a Regresson Problem { }, 3 L C K To treat classfcaton as a regresson problem we should transform the target y nto numercal values; The choce of numercal class

More information

Chapter 9: Statistical Inference and the Relationship between Two Variables

Chapter 9: Statistical Inference and the Relationship between Two Variables Chapter 9: Statstcal Inference and the Relatonshp between Two Varables Key Words The Regresson Model The Sample Regresson Equaton The Pearson Correlaton Coeffcent Learnng Outcomes After studyng ths chapter,

More information

Lecture 6: Introduction to Linear Regression

Lecture 6: Introduction to Linear Regression Lecture 6: Introducton to Lnear Regresson An Manchakul amancha@jhsph.edu 24 Aprl 27 Lnear regresson: man dea Lnear regresson can be used to study an outcome as a lnear functon of a predctor Example: 6

More information

Week 5: Neural Networks

Week 5: Neural Networks Week 5: Neural Networks Instructor: Sergey Levne Neural Networks Summary In the prevous lecture, we saw how we can construct neural networks by extendng logstc regresson. Neural networks consst of multple

More information

Chapter 3 Describing Data Using Numerical Measures

Chapter 3 Describing Data Using Numerical Measures Chapter 3 Student Lecture Notes 3-1 Chapter 3 Descrbng Data Usng Numercal Measures Fall 2006 Fundamentals of Busness Statstcs 1 Chapter Goals To establsh the usefulness of summary measures of data. The

More information

Logistic Regression. CAP 5610: Machine Learning Instructor: Guo-Jun QI

Logistic Regression. CAP 5610: Machine Learning Instructor: Guo-Jun QI Logstc Regresson CAP 561: achne Learnng Instructor: Guo-Jun QI Bayes Classfer: A Generatve model odel the posteror dstrbuton P(Y X) Estmate class-condtonal dstrbuton P(X Y) for each Y Estmate pror dstrbuton

More information

Basic R Programming: Exercises

Basic R Programming: Exercises Basc R Programmng: Exercses RProgrammng John Fox ICPSR, Summer 2009 1. Logstc Regresson: Iterated weghted least squares (IWLS) s a standard method of fttng generalzed lnear models to data. As descrbed

More information

T E C O L O T E R E S E A R C H, I N C.

T E C O L O T E R E S E A R C H, I N C. T E C O L O T E R E S E A R C H, I N C. B rdg n g En g neern g a nd Econo mcs S nce 1973 THE MINIMUM-UNBIASED-PERCENTAGE ERROR (MUPE) METHOD IN CER DEVELOPMENT Thrd Jont Annual ISPA/SCEA Internatonal Conference

More information

CS 2750 Machine Learning. Lecture 5. Density estimation. CS 2750 Machine Learning. Announcements

CS 2750 Machine Learning. Lecture 5. Density estimation. CS 2750 Machine Learning. Announcements CS 750 Machne Learnng Lecture 5 Densty estmaton Mlos Hauskrecht mlos@cs.ptt.edu 539 Sennott Square CS 750 Machne Learnng Announcements Homework Due on Wednesday before the class Reports: hand n before

More information

Chapter 12 Analysis of Covariance

Chapter 12 Analysis of Covariance Chapter Analyss of Covarance Any scentfc experment s performed to know somethng that s unknown about a group of treatments and to test certan hypothess about the correspondng treatment effect When varablty

More information

Lecture 3. Ax x i a i. i i

Lecture 3. Ax x i a i. i i 18.409 The Behavor of Algorthms n Practce 2/14/2 Lecturer: Dan Spelman Lecture 3 Scrbe: Arvnd Sankar 1 Largest sngular value In order to bound the condton number, we need an upper bound on the largest

More information

CS4495/6495 Introduction to Computer Vision. 3C-L3 Calibrating cameras

CS4495/6495 Introduction to Computer Vision. 3C-L3 Calibrating cameras CS4495/6495 Introducton to Computer Vson 3C-L3 Calbratng cameras Fnally (last tme): Camera parameters Projecton equaton the cumulatve effect of all parameters: M (3x4) f s x ' 1 0 0 0 c R 0 I T 3 3 3 x1

More information

j) = 1 (note sigma notation) ii. Continuous random variable (e.g. Normal distribution) 1. density function: f ( x) 0 and f ( x) dx = 1

j) = 1 (note sigma notation) ii. Continuous random variable (e.g. Normal distribution) 1. density function: f ( x) 0 and f ( x) dx = 1 Random varables Measure of central tendences and varablty (means and varances) Jont densty functons and ndependence Measures of assocaton (covarance and correlaton) Interestng result Condtonal dstrbutons

More information

4.3 Poisson Regression

4.3 Poisson Regression of teratvely reweghted least squares regressons (the IRLS algorthm). We do wthout gvng further detals, but nstead focus on the practcal applcaton. > glm(survval~log(weght)+age, famly="bnomal", data=baby)

More information

x i1 =1 for all i (the constant ).

x i1 =1 for all i (the constant ). Chapter 5 The Multple Regresson Model Consder an economc model where the dependent varable s a functon of K explanatory varables. The economc model has the form: y = f ( x,x,..., ) xk Approxmate ths by

More information

1 Matrix representations of canonical matrices

1 Matrix representations of canonical matrices 1 Matrx representatons of canoncal matrces 2-d rotaton around the orgn: ( ) cos θ sn θ R 0 = sn θ cos θ 3-d rotaton around the x-axs: R x = 1 0 0 0 cos θ sn θ 0 sn θ cos θ 3-d rotaton around the y-axs:

More information

Hongyi Miao, College of Science, Nanjing Forestry University, Nanjing ,China. (Received 20 June 2013, accepted 11 March 2014) I)ϕ (k)

Hongyi Miao, College of Science, Nanjing Forestry University, Nanjing ,China. (Received 20 June 2013, accepted 11 March 2014) I)ϕ (k) ISSN 1749-3889 (prnt), 1749-3897 (onlne) Internatonal Journal of Nonlnear Scence Vol.17(2014) No.2,pp.188-192 Modfed Block Jacob-Davdson Method for Solvng Large Sparse Egenproblems Hongy Mao, College of

More information

ρ some λ THE INVERSE POWER METHOD (or INVERSE ITERATION) , for , or (more usually) to

ρ some λ THE INVERSE POWER METHOD (or INVERSE ITERATION) , for , or (more usually) to THE INVERSE POWER METHOD (or INVERSE ITERATION) -- applcaton of the Power method to A some fxed constant ρ (whch s called a shft), x λ ρ If the egenpars of A are { ( λ, x ) } ( ), or (more usually) to,

More information

Matrix Approximation via Sampling, Subspace Embedding. 1 Solving Linear Systems Using SVD

Matrix Approximation via Sampling, Subspace Embedding. 1 Solving Linear Systems Using SVD Matrx Approxmaton va Samplng, Subspace Embeddng Lecturer: Anup Rao Scrbe: Rashth Sharma, Peng Zhang 0/01/016 1 Solvng Lnear Systems Usng SVD Two applcatons of SVD have been covered so far. Today we loo

More information

Chapter 13: Multiple Regression

Chapter 13: Multiple Regression Chapter 13: Multple Regresson 13.1 Developng the multple-regresson Model The general model can be descrbed as: It smplfes for two ndependent varables: The sample ft parameter b 0, b 1, and b are used to

More information