Econometrics I. Professor William Greene Stern School of Business Department of Economics 21-1/67. Part 21: Generalized Method of Moments

Size: px
Start display at page:

Download "Econometrics I. Professor William Greene Stern School of Business Department of Economics 21-1/67. Part 21: Generalized Method of Moments"

Transcription

1 Ecoometrcs I Professor Wllam Greee Ster School of Busess Departmet of Ecoomcs 21-1/67

2 Ecoometrcs I Part 21 Geeralzed Method of Momets 21-2/67

3 I also have a questos about olear GMM - whch s more or less olear IV techque I suppose. I am rug a pael o-lear regresso (o-lear the parameters) ad I have L parameters ad K exogeous varables wth L>K. I partcular my model looks kd of lke ths: Y = b 1 *X^b 2 + e, ad so I am tryg to estmate the extra b 2 that do't usually appear a regresso. From what I am readg, to ru olear GMM I ca use the K exogeous varables to costruct the orthogoalty codtos but what should I use for the extra, b 2 coeffcets? Just some more possble IVs (lke lags) of the exogeous varables? I agree that by addg more IVs you wll get a more effcet estmato, but s't t oly the case whe you beleve the IVs are truly ucorrelated wth the error term? So by addg more "strumets" you are more or less mposg more ad more restrctve assumptos about the model (whch mght ot actually be true). I am askg because I have ot foud sources comparg olear GMM/IV to olear least squares. If there s o homoscadestcty/seral correlato what s more effcet/gve tghter estmates? 21-3/67

4 I m tryg to mmze a olear program wth the least square uder olear costrats. It s frst troduced by Aé & Gema (2000). It cossted o the mmzato of the sum of squared dfferece betwee the momet geeratg fucto ad the theoretcal momet geeratg fucto (The method was suggested by Quadt ad Ramsey the 1970s.) 21-4/67

5 Method of Momet Geeratg Fuctos For the ormal dstrbuto, the MGF s M(t, )=E[exp(tx)]=exp[t + 2 t ] Momet Equatos: exp( t ) exp[t + 1 jx j 2 t j ], j 1, 2. Choose two values of t ad solve the two momet equatos for ad. Mxture of Normals Problem: f(x,,,, ) N[, ] (1 ) N[, ] Use the method of momet geeratg fuctos wth 5 values of t M(t,,,, )=E[exp(tx)]= exp[t + t ] (1 )exp[t + t ] /67

6 Fdg the solutos to the momet equatos: Least squares 1 ˆM(t 1) exp( tx 1 1 ), ad lkewse for t 2,... Mmze(,,,, ) ˆM(t ) exp[t + ] (1 )exp[t + ] j 1 2t j 1 2 2t 2 2 Alteratve estmator: Maxmum Lkelhood L(,,,, ) log N[x, ] (1 ) N[x, ] /67

7 The Method of Momets Estmatg Parameters of Dstrbutos Usg Momet Equatos Populato Momet k E[x ] f (,,..., ) k k 1 2 K Sample Momet m x. m may also be h (x ), eed ot be powers 1 k 1 k 1 k 1 k Law of Large Numbers plm m f (,,..., ) k k k 1 2 K 'Momet Equato' (k = 1,...,K) m x f (,,..., ) 1 N k k N 1 k 1 2 K Method of Momets appled by vertg the momet equatos. ˆ g (m,...,m ), k = 1,...,K k k 1 K 21-7/67

8 Estmatg a Parameter Mea of Posso p(y)=exp(-λ) λ y / y! E[y]= λ. plm (1/)Σ y = λ. Ths s the estmator Mea of Expoetal p(y) = α exp(- α y) E[y] = 1/ α. plm (1/)Σ y = 1/ α 21-8/67

9 Mea ad Varace of a Normal Dstrbuto 2 1 (y ) p(y) exp Populato Momets E[y], E[y ] Momet Equatos y, y Method of Momets Estmators ˆ=y, ˆ y (y ) (y y) /67

10 Gamma Dstrbuto p(y) E[y] P exp( y)y (P) P P1 2 P(P 1) 3 P(P 1)(P 2) E[y ] E[y ] ad so o 2 3 E[1/ y] P 1 E[logy] (P) log, (P)=dl (P)/dP (Each par gves a dfferet aswer. Is there a 'best' par? Yes, the oes that are 'suffcet' statstcs. E[y] ad E[logy]. For a dfferet course...) 21-10/67

11 The Lear Regresso Model Populato y x 1 1 Populato Expectato E[ x ] 0, k = 1,...,K k Momet Equatos (y x x... x )x K K 1 (y x x... x )x K K 2 1 (y x1 1 x x K K)xK 0 Soluto : Lear system of K equatos K ukows. Least Squares 21-11/67

12 Istrumetal Varables Populato y x 1 1 Populato Expectato E[ z ] 0 for strumetal varables z... z. k 1 K Momet Equatos (y x x... x )z K K 1 (y x x... x )z K K 2 1 (y 1 x1 1 x x KK)zK 0 Soluto : Also a lear system of K equatos K ukows. -1 b = ( Z'X / ) ( Z'y / ) IV A exteso : What s the soluto f there are M > K IVs? 21-12/67

13 Maxmum Lkelhood 1 Log lkelhood fucto, logl = logf(y x,,..., ) Populato Expectatos logl E 0, k = 1,...,K k Sample Momets 1 1 logf(y x,,..., ) 1 K k K Soluto : K olear equatos K ukows. 1 logf(y x, ˆ,..., ˆ ) 1,MLE K,MLE 1 ˆ k,mle /67

14 Behavoral Applcato Lfe Cycle Cosumpto (text, pages ) 1 c t1 E t(1 r) 1t 0 1 ct dscout rate c t cosumpto formato at tme t t Let =1/(1+ ), R c / c, =- t t1 t t+1 t1 t E [ (1 r)r 1 ] 0 What s the formato set? Each pece of 'formato' provdes a momet equato for estmato of the two parameters. (1 r) 1 w 0, k=1,...,k T 1 ct 1 t1 tk 1 c t 21-14/67

15 Idetfcato Ca the parameters be estmated? Not a sample property Assume a fte sample Is there suffcet formato a sample to reveal cosstet estmators of the parameters Ca the momet equatos be solved for the populato parameters? 21-15/67

16 Idetfcato Exactly Idetfed Case: K populato momet equatos K ukow parameters. Our famlar cases, OLS, IV, ML, the MOM estmators Is the coutg rule suffcet? What else s eeded? Overdetfed Case Istrumetal Varables Uderdetfed Case Multcollearty Varace parameter a probt model Shape parameter a loglear model 21-16/67

17 Populato y x,,..., 1 K Populato Expectato 1 Overdetfcato E[ z ] 0 for strumetal varables z... z M > K. k 1 M There are M > K Momet Equatos - more tha ecessary (y x x... x )z K K 1 (y x x... x )z K K 2 1 (y x1 1 x x K K)zM 0 Soluto : A lear system of M equatos K ukows /67

18 Overdetfcato Two Equato Covarace Structures Model Coutry 1: Coutry 2: y y Two Populato Momet Codtos: E[(1/T) X '( y X )] E[(1/T) X '( y X )] X X (1) How do we combe the two sets of equatos? (2) Gve two OLS estmates, b ad b, how do we recocle them? 1 2 Note: There are eve more. E[(1/T) X '( y X )] /67

19 Uderdetfcato Model/Data Cosder the Mover - Stayer Model Bary choce for whether a dvdual 'moves' or 'stays' d 1( z u 0) Outcome equato for the dvdual, codtoal o the state: y (d 0) = y (d 1) x 0 0 = x (, ) ~ N[(0,0),(,, )] A dvdual ether moves or stays, but ot both (or ether). The parameter caot be estmated wth the observed data regardless of the sample sze. It s udetfed /67

20 Uderdetfcato - Normalzato Whe a parameter s udetfed, the log lkelhood s varat to chages t. Cosder the logt bary choce model exp( 0x) exp( 1x) Prob[y=0]= Prob[y=1]= exp( x) exp( x) exp( x) exp( x) Probabltes sum to 1, are mootoc, etc. But, cosder, for ay 0, exp[( 0 )x] exp( x) [exp( 0x)] Prob[y=0]= exp[( )x] exp[( )x] exp( x)[exp( x) exp( x)] exp[( 1 )x] exp( x) [exp( 1x)] Prob[y=1]= exp[( )x] exp [( )x] exp( x)[exp( x) exp( x)] exp( x) always cacels out. The parameters are udetfed. A ormalzato such as 0 s eeded /67

21 Uderdetfcato: Momets Nolear LS vs. MLE y ~ Gamma(P, ), exp( x ) f (y ) P E[y x] exp( y ) y ( P) P P1 We cosder olear least squares ad maxmum lkelhood estmato of the parameters. We use the Germa health care data, where y = come x = 1,age,educ,female,hhkds,marred 21-21/67

22 21-22/67 Nolear Least Squares --> NAMELIST ; x = oe,age,educ,female,hhkds,marred $ --> Calc ; k=col(x) $ --> NLSQ ; Lhs = hhc ; Fc = p / exp(b1'x) ; labels = k_b,p ; start = k_0,1 ; maxt = 20$ Momet matrx has become opostve defte. Swtchg to BFGS algorthm Normal ext: 16 teratos. Status=0. F= User Defed Optmzato... Nolear least squares regresso... LHS=HHNINC Mea = Stadard devato = Number of observs. = Model sze Parameters = 7 Degrees of freedom = Resduals Sum of squares = Stadard error of e = Varable Coeffcet Stadard Error b/st.er. P[ Z >z] B <====== B B *** B *** B *** B *** P <======= Nolear least squares dd ot work. That s the mplcato of the fte stadard errors for B1 (the costat) ad P.

23 Maxmum Lkelhood Gamma (Loglear) Regresso Model Depedet varable HHNINC Log lkelhood fucto Restrcted log lkelhood Ch squared [ 6 d.f.] Sgfcace level McFadde Pseudo R-squared (4 observatos wth come = 0 Estmato based o N = 27322, K = 7 were deleted so logl was computable.) Varable Coeffcet Stadard Error b/st.er. P[ Z >z] Mea of X Parameters codtoal mea fucto Costat *** AGE.00205*** EDUC *** FEMALE HHKIDS.06512*** MARRIED *** Scale parameter for gamma model P_scale *** MLE apparetly worked fe. Why dd oe method (ls) fal ad aother cosstet estmator work wthout dffculty? 21-23/67

24 E[ y x] P / exp( x ) Momet Equatos: NLS 2 y P x e e' e / exp( ) ee ' P ee ' e exp( x ) 0 2eP x 0 exp( x ) Cosder the term for the costat the model,. Notce that the frst order codto for the costat term s 1 2eP exp( x ) 0. Ths does't deped o P, sce we ca dvde both sdes of the equato by P. Ths meas that we caot fd solutos for both 1 ad P. It s easy to see why NLS caot dstgush P from. E[y x] = exp((logp- )...). There are a fte umber 1 1 of pars of (P, ) that produce the same costat term the model /67

25 1 Momet Equatos MLE The log lkelhood fucto ad lkelhood equatos are logl= P log log ( P) y ( P 1) log y log L d log ( P) log ( ) log 0, ( ) 1 P y P P dp log L P ; usg. 1 y x Recall, the expected values of the dervatves of the log lkelhood equal zero. So, a look at the frst equato reveals that the momet equato use for estmatg P s E[logy x ] ( P) log ad aother K momet P equatos, E y x 0 are also use. So, the MLE uses K+1 fuctoally depedet momet equatos for K+1 parameters, whle NLS was oly usg K depedet momet equatos for the same K+1 parameters /67

26 GMM Ageda The Method of Momets. Solvg the momet equatos Exactly detfed cases Overdetfed cases Cosstecy. How do we kow the method of momets s cosstet? Asymptotc covarace matrx. Cosstet vs. Effcet estmato A weghtg matrx The mmum dstace estmator What s the effcet weghtg matrx? Estmatg the weghtg matrx. The Geeralzed method of momets estmator - how t s computed. Computg the approprate asymptotc covarace matrx 21-26/67

27 The Method of Momets Momet Equato: Defes a sample statstc that mmcs a populato expectato: The populato expectato orthogoalty codto: E[ m () ] = 0. Subscrpt dcates t depeds o data vector dexed by '' (or 't' for a tme seres settg) 21-27/67

28 The Method of Momets - Example Gamma Dstrbuto Parameters p(y ) exp( y )y P P1 (P) Populato Momet Codtos P E[y ], E[logy ] (P) log Momet Equatos: E[m (,P)] = E[{(1/) y } P / ] 0 1 =1 E[m (,P)] = E[{(1/) logy } ((P) log )] 0 2 = /67

29 PSI Applcato Solvg the momet equatos Use least squares: Mmze {m E[m ]} {m E[m ]} (m (P / )) (m ( (P) log )) m m Plot of Ps(P) Fucto P 21-29/67

30 Method of Momets Soluto create ; y1=y ; y2=log(y)$ calc ; m1=xbr(y1) ; ms=xbr(y2)$ mmze; start = 2.0,.06 ; labels = p,l ; fc = (l*m1-p)^2 + (ms - ps(p)+log(l)) ^2 $ User Defed Optmzato Depedet varable Fucto Number of observatos 1 Iteratos completed 6 Log lkelhood fucto e Varable Coeffcet P L /67

31 Nolear Istrumetal Varables There are K parameters, y = f(x,) +. There exsts a set of K strumetal varables, z such that E[z ] = 0. The sample couterpart s the momet equato (1/) z = (1/) z [y - f(x,)] = (1/) m () = m() = 0. The method of momets estmator s the soluto to the momet equato(s). (How the soluto s obtaed s ot always obvous, ad vares from problem to problem.) 21-31/67

32 The MOM Soluto There are K equatos K ukows m( )= 0 If there s a soluto, there s a exact soluto At the soluto, m( )= 0, ad [ m( )]'[ m( )] = 0 Sce [ m( )]'[ m( )] 0, the soluto ca be foud by solvg the programmg problem Mmze wrt : [ m( )]'[ m( )] For ths problem, [ m( )]'[ m( )] = [(1/) ε'z] [(1/) Z'ε] The soluto s defed by [ m( )]'[ m( )] [(1/) ε'z ] [(1/) Z'ε] = 21-32/67

33 MOM Soluto [(1/) ε'z] [(1/) Z'ε] 2 (1/ ) G'Z [(1/) Z'ε] G = K matrx wth row equal to g f( x, ) For the classcal lear regresso model, f( x, ) x ' Z = X, G = X, ad the FOC are -2[(1/)( X'X)] [(1/) X' ε] = 0 ˆ -1 whch has uque soluto =( X'X) X'y 21-33/67

34 Varace of the Method of Momets Estmator The MOM estmator solves m( β)= 0 1 m β 1 ( )= m β Ω Ω ( ) so the varace s for some 1 Geerally, Ω = E[ m ( β) m ( β )'] Asy.Var[ βmom]=( G) Ω( G' ) where The asymptotc covarace matrx of the estmator s G= m( β) β' 21-34/67

35 Example 1: Gamma Dstrbuto m (y ) 1 P 1 1 m (log y (P) log ) Var(y ) Cov(y,logy ) Cov(y,logy ) Var(logy ) 1 P 1 N 2 G 1 1 '(P) 1 y y ˆ y log log 1 y y y log y log y 21-35/67

36 Example 2: Nolear IV Least Squares y f( x, ), z = the set of K strumetal varables Var[ ] m z Var[ m ] z z ' Var[ m( )]=(1/ ) z z ' ( / ) Z' Z 0 G (1/ ) z x ' 2 2 Wth depedet observatos, observatos are ucorrelated x 1 f( x, ). I the lear model, ths s just x. 0 0 G (1/ ) Z' X. 0 where x 1 ( G ) V( G )' [ (1/ ) Z' X ] [( / ) Z' Z][ (1/ ) X ' Z] = [ Z' X ] [ Z' Z][ X ' Z] s the vector of 'pseudo-regressors,' 21-36/67

37 Varace of the Momets How to estmate V = (1/) = Var[ m( )] Var[ m( )]=(1/)Var[ m( )] = (1/) Vˆ (1/ ) (1/ ) m ( ˆ) m ( ˆ)' Estmate Var[ m ( )] wth Est.Var[ m ( )] = (1/) m ( ) m ( )' The, =1 =1 For the lear regresso model, m x, Vˆ 2 (1/ ) (1/ ) x e e x ' (1/ ) (1/ ) e x x ' G (1/ ) X'X Est.Var[ b ] [(1/ ) X'X] [(1/ ) (1/ ) e x x '][(1/ ) X'X] = [ X'X] [ e x x '][ X'X] =1 = MOM = =1 (famlar?) 21-37/67

38 Cosstet? Propertes of the MOM Estmator The LLN mples that the momets are cosstet estmators of ther populato couterparts (zero) Use the Slutsky theorem to assert cosstecy of the fuctos of the momets Asymptotcally ormal? The momets are sample meas. Ivoke a cetral lmt theorem. Effcet? Not ecessarly Sometmes yes. (Gamma example) Perhaps ot. Depeds o the model ad the avalable formato (ad how much of t s used) /67

39 Geeralzg the Method of Momets Estmator More momets tha parameters the overdetfed case Example: Istrumetal varable case, M > K strumets 21-39/67

40 Two Stage Least Squares How to use a excess of strumetal varables (1) X s K varables. Some (at least oe) of the K varables X are correlated wth ε. (2) Z s M > K varables. Some of the varables Z are also X, some are ot. Noe of the varables Z are correlated wth ε. (3) Whch K varables to use to compute Z X ad Z y? 21-40/67

41 Choose K radomly? Choosg the Istrumets Choose the cluded Xs ad the remader radomly? Use all of them? How? A theorem: (Brudy ad Jorgeso, ca. 1972) There s a most effcet way to costruct the IV estmator from ths subset: (1) For each colum (varable) X, compute the predctos of that varable usg all the colums of Z. (2) Learly regress y o these K predctos. Ths s two stage least squares 21-41/67

42 2SLS Algebra Xˆ -1 Z(Z'Z) Z'X ˆ ˆ ˆ 1 b2sls ( X'X) X'y -1 But, Z(Z'Z) Z'X = ( I - M ) X ad ( I - M ) s dempotet. X'X ˆˆ = X' ( I - M )( I - M ) X = X' ( I - M ) X so b X'X ˆ X'y ˆ 2SLS 1 ( ) = a real IV estmator by the defto. Note, plm( X' ˆ /) = 0 sce colums of Z Z Z Z Z Xˆ are lear combatos of the colums of Z, all of whch are ucorrelated wth b X' I - M X X' I - M y -1 2SLS ( Z) ] ( Z) 21-42/67

43 Method of Momets Estmato Same Momet Equato m( β)= 0 Now, M momet equatos, K parameters. There s o uque soluto. There s also o exact soluto to m( β)= 0. We get as close as we ca. How to choose the estmator? Least squares s a obvous choce. Mmze wrt β : m( β )'m( β) E.g., Mmze wrt β : [(1/) ( β )'Z][(1/) Z' ( β)]=(1/ 2 ) ( β )'ZZ' ( β) 21-43/67

44 FOC for MOM Frst order codtos (1) Geeral m( β )'m( β)/ β = 2 G( β )'m( β) = 0 (2) The Istrumetal Varables Problem 2 2 (1/ ) ( β )'ZZ' ( β)/ β = - (2/ )( X ' Z)[ Z' ( y - Xβ)] = 0 Or, ( X ' Z)[ Z' ( y - Xβ)] = 0 (K M) (M )( 1) = 0 At the soluto, ( X ' Z)[ Z' ( y - Xβ)] = 0 But, [ Z' ( y - Xβ)] 0 as t was before /67

45 Computg the Estmator Programmg Program No all purpose soluto Nolear optmzato problem soluto vares from settg to settg /67

46 Asymptotc Covarace Matrx Geeral Result for Method of Momets whe M K Momet Equatos:E[ m( β)]= 0 Soluto - FOC: G( β )'m( β)= 0, G( β )' s K M Asymptotc Covarace Matrx ˆ -1-1 Asy.Var[ β] = [ G( β )' V G( β)], V = Asy.Var[ m β Specal Case - Exactly Idetfed: M = K ad -1 G( β) s osgular. The [ G( β)] exsts ad Asy.Var[ βˆ ] = [ G( β)] V [ G( β )'] -1-1 ( )] 21-46/67

47 MD More Effcet Estmato We have used least squares, Mmze wrt β : m( β )'m( β) to fd the estmator of β. Is ths the most effcet way to proceed? Geerally ot: We cosder a more geeral approach Mmum Dstace Estmato Let A be ay postve defte matrx: Let βˆ = the soluto to Mmze wrt β : q = m( β )' A m( β) Ths s a mmum dstace ( the metrc of A) estmator /67

48 MD Mmum Dstace Estmato Let A be ay postve defte matrx: Let βˆ = the soluto to Mmze wrt β : q = m( β )' A m( β) where E[ m( β)] = 0 (the usual momet codtos). Ths s a mmum dstace ( the metrc of A) estmator. βˆ s cosstet MD βˆ s asymptotcally ormally dstrbuted. MD Same argumets as for the GMM estmator. Effcecy of the estmator depeds o the choce of A /67

49 MDE Estmato: Applcato N uts, T observatos per ut, T > K y Xβ ε, E[ ε X ] 0 Cosder the followg estmato strategy: (1) OLS coutry by coutry, b produces N estmators of (2) How to combe the estmators? b1 β We hav b2 β e 'momet' equato: E 0... bn β How ca I combe the N estmators of β? β 21-49/67

50 Least Squares b1 β b1 β b2 β b2 β E 0. m( β)= bn β bn β To mmze m( β)' m( β) = ( b β )'( b β) b1 β m( β)' m( β) b2 β N 2[ I, I,..., I] 2 ( b β 0 ). 1 β... bn β N 1 N The soluto s ( b β) 0 or β = b b 1 1 N N /67

51 21-51/67 Geeralzed Least Squares The precedg used OLS - smple uweghted least squares. I A 0 I... 0 Also, t uses =. Suppose we use weghted, GLS I [ XX 1 ( 1 1) ] [ XX 2( 2 2 The, A = ) ] [ XX N( N N) ] The frst order codto for mmzg m( β)' Am( β) s N X X b β =1 {[ ( ) ] }( ) = N N XX =1 XX) ] } b =1 or β = {[ ( ) ] } {[ ( N = Wb = a matrx weghted average. =1

52 Mmum Dstace Estmato The mmum dstace estmator mmzes q = m( β )' A m( β) The estmator s (1) Cosstet (2) Asymptotcally ormally dstrbuted (3) Has asymptotc covarace matrx Asy.Var[ ˆ β ] [ G( β) 'AG( β)] [ G( β )'AVAG( β)][ G( β )'AG( β)] MD /67

53 Optmal Weghtg Matrx A s the Weghtg Matrx of the mmum dstace estmator. Are some A's better tha others? (Yes) Is there a best choce for A? Yes The varace of the MDE s mmzed whe A = {Asy. Var[ m( β)]} Ths defes the geeralzed method of momets estmator. A = V /67

54 GMM Estmato 1 m( β)= 1m (y, x,β) 11 Asy.Var[ m( β)] estmated wth W= 1m (y, x,β) m (y, x,β) The GMM estmator of β the mmzes q 1m (y, x,β ) 'W 1m (y, x,β) m( β) Est.Asy.Var[ˆ βgm M] [ G'W G], G = β 21-54/67

55 GMM Estmato Exactly detfed GMM problems 1 Whe m( β) = 1m (y, x,β) 0 s K equatos K ukow parameters (the exactly detfed case), the weghtg matrx q 1m (y, x, β ) 'W 1m (y, x,β) s rrelevat to the soluto, sce we ca set exactly m( β) 0 so q = 0. Ad, the asymptotc covarace matrx (estmator) s the product of 3 square matrces ad becomes [ G'W G] G WG' /67

56 A Practcal Problem Asy.Var[ m( β)] estmated wth 11 W= 1m (y, x, β) m (y, x,β) The GMM estmator of β the mmzes q 1m (y, x, β ) 'W 1m (y, x,β). I order to compute W, you eed to kow β, ad you are tryg to estmate β. How to proceed? Typcally two steps: (1) Use A = I. Smple least squares, to get a prelmary estmator of β. Ths s cosstet, though ot effcet. (2) Compute the weghtg matrx, the use GMM /67

57 Iferece Testg hypotheses about the parameters: Wald test A couterpart to the lkelhood rato test Testg the overdetfyg restrctos 21-57/67

58 Testg Hypotheses (1) Wald Tests the usual fasho. (2) A couterpart to lkelhood rato tests GMM crtero s q = m( β )'W m( β) whe restrctos are mposed o β q creases. d qrestrcted qurestrcted ch squared[j] (The weghtg matrx must be the same for both.) (3) Testg the overdetfyg restrctos: q would be 0 f exactly detfed. q - 0 > 0 results from the overdetfyg restrctos /67

59 Applcato: Iovato 21-59/67

60 Applcato: Iovato 21-60/67

61 Applcato: Multvarate Probt Model 5 - varate Probt Model y * x, y 1[y * 0] t t t t t x x x x x log [{(2 1), 1,...,5}, ] L y s t ds ds ds ds ds 5 t t Requres 5 dmesoal tegrato of the jot ormal desty. Very hard! But, E[y x ] ( x ). t t t Orthogoalty Codtos: E[{y - ( x )} x 0 t t t {y - ( x )} x {y - ( )} 1 x x Momet Equatos: {y - ( x )} x 1 {y - ( )} {y - ( x )} x x4 x equatos 8 parameters /67

62 Pooled Probt Igorg Correlato 21-62/67

63 Radom Effects: Σ=(1- ρ)i+ρ 21-63/67

64 Urestrcted Correlato Matrx 21-64/67

ECON 5360 Class Notes GMM

ECON 5360 Class Notes GMM ECON 560 Class Notes GMM Geeralzed Method of Momets (GMM) I beg by outlg the classcal method of momets techque (Fsher, 95) ad the proceed to geeralzed method of momets (Hase, 98).. radtoal Method of Momets

More information

Wu-Hausman Test: But if X and ε are independent, βˆ. ECON 324 Page 1

Wu-Hausman Test: But if X and ε are independent, βˆ. ECON 324 Page 1 Wu-Hausma Test: Detectg Falure of E( ε X ) Caot drectly test ths assumpto because lack ubased estmator of ε ad the OLS resduals wll be orthogoal to X, by costructo as ca be see from the momet codto X'

More information

Econometric Methods. Review of Estimation

Econometric Methods. Review of Estimation Ecoometrc Methods Revew of Estmato Estmatg the populato mea Radom samplg Pot ad terval estmators Lear estmators Ubased estmators Lear Ubased Estmators (LUEs) Effcecy (mmum varace) ad Best Lear Ubased Estmators

More information

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS Postpoed exam: ECON430 Statstcs Date of exam: Jauary 0, 0 Tme for exam: 09:00 a.m. :00 oo The problem set covers 5 pages Resources allowed: All wrtte ad prted

More information

( ) = ( ) ( ) Chapter 13 Asymptotic Theory and Stochastic Regressors. Stochastic regressors model

( ) = ( ) ( ) Chapter 13 Asymptotic Theory and Stochastic Regressors. Stochastic regressors model Chapter 3 Asmptotc Theor ad Stochastc Regressors The ature of eplaator varable s assumed to be o-stochastc or fed repeated samples a regresso aalss Such a assumpto s approprate for those epermets whch

More information

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS Exam: ECON430 Statstcs Date of exam: Frday, December 8, 07 Grades are gve: Jauary 4, 08 Tme for exam: 0900 am 00 oo The problem set covers 5 pages Resources allowed:

More information

Special Instructions / Useful Data

Special Instructions / Useful Data JAM 6 Set of all real umbers P A..d. B, p Posso Specal Istructos / Useful Data x,, :,,, x x Probablty of a evet A Idepedetly ad detcally dstrbuted Bomal dstrbuto wth parameters ad p Posso dstrbuto wth

More information

Ordinary Least Squares Regression. Simple Regression. Algebra and Assumptions.

Ordinary Least Squares Regression. Simple Regression. Algebra and Assumptions. Ordary Least Squares egresso. Smple egresso. Algebra ad Assumptos. I ths part of the course we are gog to study a techque for aalysg the lear relatoshp betwee two varables Y ad X. We have pars of observatos

More information

Lecture 7. Confidence Intervals and Hypothesis Tests in the Simple CLR Model

Lecture 7. Confidence Intervals and Hypothesis Tests in the Simple CLR Model Lecture 7. Cofdece Itervals ad Hypothess Tests the Smple CLR Model I lecture 6 we troduced the Classcal Lear Regresso (CLR) model that s the radom expermet of whch the data Y,,, K, are the outcomes. The

More information

Chapter 14 Logistic Regression Models

Chapter 14 Logistic Regression Models Chapter 4 Logstc Regresso Models I the lear regresso model X β + ε, there are two types of varables explaatory varables X, X,, X k ad study varable y These varables ca be measured o a cotuous scale as

More information

Linear Regression with One Regressor

Linear Regression with One Regressor Lear Regresso wth Oe Regressor AIM QA.7. Expla how regresso aalyss ecoometrcs measures the relatoshp betwee depedet ad depedet varables. A regresso aalyss has the goal of measurg how chages oe varable,

More information

Chapter 5 Properties of a Random Sample

Chapter 5 Properties of a Random Sample Lecture 6 o BST 63: Statstcal Theory I Ku Zhag, /0/008 Revew for the prevous lecture Cocepts: t-dstrbuto, F-dstrbuto Theorems: Dstrbutos of sample mea ad sample varace, relatoshp betwee sample mea ad sample

More information

Point Estimation: definition of estimators

Point Estimation: definition of estimators Pot Estmato: defto of estmators Pot estmator: ay fucto W (X,..., X ) of a data sample. The exercse of pot estmato s to use partcular fuctos of the data order to estmate certa ukow populato parameters.

More information

TESTS BASED ON MAXIMUM LIKELIHOOD

TESTS BASED ON MAXIMUM LIKELIHOOD ESE 5 Toy E. Smth. The Basc Example. TESTS BASED ON MAXIMUM LIKELIHOOD To llustrate the propertes of maxmum lkelhood estmates ad tests, we cosder the smplest possble case of estmatg the mea of the ormal

More information

Chapter 4 Multiple Random Variables

Chapter 4 Multiple Random Variables Revew for the prevous lecture: Theorems ad Examples: How to obta the pmf (pdf) of U = g (, Y) ad V = g (, Y) Chapter 4 Multple Radom Varables Chapter 44 Herarchcal Models ad Mxture Dstrbutos Examples:

More information

STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS. x, where. = y - ˆ " 1

STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS. x, where. = y - ˆ  1 STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS Recall Assumpto E(Y x) η 0 + η x (lear codtoal mea fucto) Data (x, y ), (x 2, y 2 ),, (x, y ) Least squares estmator ˆ E (Y x) ˆ " 0 + ˆ " x, where ˆ

More information

Recall MLR 5 Homskedasticity error u has the same variance given any values of the explanatory variables Var(u x1,...,xk) = 2 or E(UU ) = 2 I

Recall MLR 5 Homskedasticity error u has the same variance given any values of the explanatory variables Var(u x1,...,xk) = 2 or E(UU ) = 2 I Chapter 8 Heterosedastcty Recall MLR 5 Homsedastcty error u has the same varace gve ay values of the eplaatory varables Varu,..., = or EUU = I Suppose other GM assumptos hold but have heterosedastcty.

More information

ECONOMETRIC THEORY. MODULE VIII Lecture - 26 Heteroskedasticity

ECONOMETRIC THEORY. MODULE VIII Lecture - 26 Heteroskedasticity ECONOMETRIC THEORY MODULE VIII Lecture - 6 Heteroskedastcty Dr. Shalabh Departmet of Mathematcs ad Statstcs Ida Isttute of Techology Kapur . Breusch Paga test Ths test ca be appled whe the replcated data

More information

ENGI 3423 Simple Linear Regression Page 12-01

ENGI 3423 Simple Linear Regression Page 12-01 ENGI 343 mple Lear Regresso Page - mple Lear Regresso ometmes a expermet s set up where the expermeter has cotrol over the values of oe or more varables X ad measures the resultg values of aother varable

More information

X X X E[ ] E X E X. is the ()m n where the ( i,)th. j element is the mean of the ( i,)th., then

X X X E[ ] E X E X. is the ()m n where the ( i,)th. j element is the mean of the ( i,)th., then Secto 5 Vectors of Radom Varables Whe workg wth several radom varables,,..., to arrage them vector form x, t s ofte coveet We ca the make use of matrx algebra to help us orgaze ad mapulate large umbers

More information

Lecture 3. Sampling, sampling distributions, and parameter estimation

Lecture 3. Sampling, sampling distributions, and parameter estimation Lecture 3 Samplg, samplg dstrbutos, ad parameter estmato Samplg Defto Populato s defed as the collecto of all the possble observatos of terest. The collecto of observatos we take from the populato s called

More information

X ε ) = 0, or equivalently, lim

X ε ) = 0, or equivalently, lim Revew for the prevous lecture Cocepts: order statstcs Theorems: Dstrbutos of order statstcs Examples: How to get the dstrbuto of order statstcs Chapter 5 Propertes of a Radom Sample Secto 55 Covergece

More information

Multivariate Transformation of Variables and Maximum Likelihood Estimation

Multivariate Transformation of Variables and Maximum Likelihood Estimation Marquette Uversty Multvarate Trasformato of Varables ad Maxmum Lkelhood Estmato Dael B. Rowe, Ph.D. Assocate Professor Departmet of Mathematcs, Statstcs, ad Computer Scece Copyrght 03 by Marquette Uversty

More information

Summary of the lecture in Biostatistics

Summary of the lecture in Biostatistics Summary of the lecture Bostatstcs Probablty Desty Fucto For a cotuos radom varable, a probablty desty fucto s a fucto such that: 0 dx a b) b a dx A probablty desty fucto provdes a smple descrpto of the

More information

Lecture 3 Probability review (cont d)

Lecture 3 Probability review (cont d) STATS 00: Itroducto to Statstcal Iferece Autum 06 Lecture 3 Probablty revew (cot d) 3. Jot dstrbutos If radom varables X,..., X k are depedet, the ther dstrbuto may be specfed by specfyg the dvdual dstrbuto

More information

Econometrics. 3) Statistical properties of the OLS estimator

Econometrics. 3) Statistical properties of the OLS estimator 30C0000 Ecoometrcs 3) Statstcal propertes of the OLS estmator Tmo Kuosmae Professor, Ph.D. http://omepre.et/dex.php/tmokuosmae Today s topcs Whch assumptos are eeded for OLS to work? Statstcal propertes

More information

GENERALIZED METHOD OF MOMENTS CHARACTERISTICS AND ITS APPLICATION ON PANELDATA

GENERALIZED METHOD OF MOMENTS CHARACTERISTICS AND ITS APPLICATION ON PANELDATA Sc.It.(Lahore),26(3),985-990,2014 ISSN 1013-5316; CODEN: SINTE 8 GENERALIZED METHOD OF MOMENTS CHARACTERISTICS AND ITS APPLICATION ON PANELDATA Beradhta H. S. Utam 1, Warsoo 1, Da Kurasar 1, Mustofa Usma

More information

Simple Linear Regression

Simple Linear Regression Statstcal Methods I (EST 75) Page 139 Smple Lear Regresso Smple regresso applcatos are used to ft a model descrbg a lear relatoshp betwee two varables. The aspects of least squares regresso ad correlato

More information

Objectives of Multiple Regression

Objectives of Multiple Regression Obectves of Multple Regresso Establsh the lear equato that best predcts values of a depedet varable Y usg more tha oe eplaator varable from a large set of potetal predctors {,,... k }. Fd that subset of

More information

Lecture Notes Types of economic variables

Lecture Notes Types of economic variables Lecture Notes 3 1. Types of ecoomc varables () Cotuous varable takes o a cotuum the sample space, such as all pots o a le or all real umbers Example: GDP, Polluto cocetrato, etc. () Dscrete varables fte

More information

Part 4b Asymptotic Results for MRR2 using PRESS. Recall that the PRESS statistic is a special type of cross validation procedure (see Allen (1971))

Part 4b Asymptotic Results for MRR2 using PRESS. Recall that the PRESS statistic is a special type of cross validation procedure (see Allen (1971)) art 4b Asymptotc Results for MRR usg RESS Recall that the RESS statstc s a specal type of cross valdato procedure (see Alle (97)) partcular to the regresso problem ad volves fdg Y $,, the estmate at the

More information

STK4011 and STK9011 Autumn 2016

STK4011 and STK9011 Autumn 2016 STK4 ad STK9 Autum 6 Pot estmato Covers (most of the followg materal from chapter 7: Secto 7.: pages 3-3 Secto 7..: pages 3-33 Secto 7..: pages 35-3 Secto 7..3: pages 34-35 Secto 7.3.: pages 33-33 Secto

More information

9.1 Introduction to the probit and logit models

9.1 Introduction to the probit and logit models EC3000 Ecoometrcs Lecture 9 Probt & Logt Aalss 9. Itroducto to the probt ad logt models 9. The logt model 9.3 The probt model Appedx 9. Itroducto to the probt ad logt models These models are used regressos

More information

COV. Violation of constant variance of ε i s but they are still independent. The error term (ε) is said to be heteroscedastic.

COV. Violation of constant variance of ε i s but they are still independent. The error term (ε) is said to be heteroscedastic. c Pogsa Porchawseskul, Faculty of Ecoomcs, Chulalogkor Uversty olato of costat varace of s but they are stll depedet. C,, he error term s sad to be heteroscedastc. c Pogsa Porchawseskul, Faculty of Ecoomcs,

More information

Qualifying Exam Statistical Theory Problem Solutions August 2005

Qualifying Exam Statistical Theory Problem Solutions August 2005 Qualfyg Exam Statstcal Theory Problem Solutos August 5. Let X, X,..., X be d uform U(,),

More information

12.2 Estimating Model parameters Assumptions: ox and y are related according to the simple linear regression model

12.2 Estimating Model parameters Assumptions: ox and y are related according to the simple linear regression model 1. Estmatg Model parameters Assumptos: ox ad y are related accordg to the smple lear regresso model (The lear regresso model s the model that says that x ad y are related a lear fasho, but the observed

More information

Answer key to problem set # 2 ECON 342 J. Marcelo Ochoa Spring, 2009

Answer key to problem set # 2 ECON 342 J. Marcelo Ochoa Spring, 2009 Aswer key to problem set # ECON 34 J. Marcelo Ochoa Sprg, 009 Problem. For T cosder the stadard pael data model: y t x t β + α + ǫ t a Numercally compare the fxed effect ad frst dfferece estmates. b Compare

More information

Multiple Regression. More than 2 variables! Grade on Final. Multiple Regression 11/21/2012. Exam 2 Grades. Exam 2 Re-grades

Multiple Regression. More than 2 variables! Grade on Final. Multiple Regression 11/21/2012. Exam 2 Grades. Exam 2 Re-grades STAT 101 Dr. Kar Lock Morga 11/20/12 Exam 2 Grades Multple Regresso SECTIONS 9.2, 10.1, 10.2 Multple explaatory varables (10.1) Parttog varablty R 2, ANOVA (9.2) Codtos resdual plot (10.2) Trasformatos

More information

STA302/1001-Fall 2008 Midterm Test October 21, 2008

STA302/1001-Fall 2008 Midterm Test October 21, 2008 STA3/-Fall 8 Mdterm Test October, 8 Last Name: Frst Name: Studet Number: Erolled (Crcle oe) STA3 STA INSTRUCTIONS Tme allowed: hour 45 mutes Ads allowed: A o-programmable calculator A table of values from

More information

Multiple Choice Test. Chapter Adequacy of Models for Regression

Multiple Choice Test. Chapter Adequacy of Models for Regression Multple Choce Test Chapter 06.0 Adequac of Models for Regresso. For a lear regresso model to be cosdered adequate, the percetage of scaled resduals that eed to be the rage [-,] s greater tha or equal to

More information

Functions of Random Variables

Functions of Random Variables Fuctos of Radom Varables Chapter Fve Fuctos of Radom Varables 5. Itroducto A geeral egeerg aalyss model s show Fg. 5.. The model output (respose) cotas the performaces of a system or product, such as weght,

More information

Maximum Likelihood Estimation

Maximum Likelihood Estimation Marquette Uverst Maxmum Lkelhood Estmato Dael B. Rowe, Ph.D. Professor Departmet of Mathematcs, Statstcs, ad Computer Scece Coprght 08 b Marquette Uverst Maxmum Lkelhood Estmato We have bee sag that ~

More information

4. Standard Regression Model and Spatial Dependence Tests

4. Standard Regression Model and Spatial Dependence Tests 4. Stadard Regresso Model ad Spatal Depedece Tests Stadard regresso aalss fals the presece of spatal effects. I case of spatal depedeces ad/or spatal heterogeet a stadard regresso model wll be msspecfed.

More information

ε. Therefore, the estimate

ε. Therefore, the estimate Suggested Aswers, Problem Set 3 ECON 333 Da Hugerma. Ths s ot a very good dea. We kow from the secod FOC problem b) that ( ) SSE / = y x x = ( ) Whch ca be reduced to read y x x = ε x = ( ) The OLS model

More information

ENGI 4421 Joint Probability Distributions Page Joint Probability Distributions [Navidi sections 2.5 and 2.6; Devore sections

ENGI 4421 Joint Probability Distributions Page Joint Probability Distributions [Navidi sections 2.5 and 2.6; Devore sections ENGI 441 Jot Probablty Dstrbutos Page 7-01 Jot Probablty Dstrbutos [Navd sectos.5 ad.6; Devore sectos 5.1-5.] The jot probablty mass fucto of two dscrete radom quattes, s, P ad p x y x y The margal probablty

More information

Lecture Note to Rice Chapter 8

Lecture Note to Rice Chapter 8 ECON 430 HG revsed Nov 06 Lecture Note to Rce Chapter 8 Radom matrces Let Y, =,,, m, =,,, be radom varables (r.v. s). The matrx Y Y Y Y Y Y Y Y Y Y = m m m s called a radom matrx ( wth a ot m-dmesoal dstrbuto,

More information

Likelihood Ratio, Wald, and Lagrange Multiplier (Score) Tests. Soccer Goals in European Premier Leagues

Likelihood Ratio, Wald, and Lagrange Multiplier (Score) Tests. Soccer Goals in European Premier Leagues Lkelhood Rato, Wald, ad Lagrage Multpler (Score) Tests Soccer Goals Europea Premer Leagues - 4 Statstcal Testg Prcples Goal: Test a Hpothess cocerg parameter value(s) a larger populato (or ature), based

More information

THE ROYAL STATISTICAL SOCIETY 2016 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE MODULE 5

THE ROYAL STATISTICAL SOCIETY 2016 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE MODULE 5 THE ROYAL STATISTICAL SOCIETY 06 EAMINATIONS SOLUTIONS HIGHER CERTIFICATE MODULE 5 The Socety s provdg these solutos to assst cadtes preparg for the examatos 07. The solutos are teded as learg ads ad should

More information

Feature Selection: Part 2. 1 Greedy Algorithms (continued from the last lecture)

Feature Selection: Part 2. 1 Greedy Algorithms (continued from the last lecture) CSE 546: Mache Learg Lecture 6 Feature Selecto: Part 2 Istructor: Sham Kakade Greedy Algorthms (cotued from the last lecture) There are varety of greedy algorthms ad umerous amg covetos for these algorthms.

More information

2SLS Estimates ECON In this case, begin with the assumption that E[ i

2SLS Estimates ECON In this case, begin with the assumption that E[ i SLS Estmates ECON 3033 Bll Evas Fall 05 Two-Stage Least Squares (SLS Cosder a stadard lear bvarate regresso model y 0 x. I ths case, beg wth the assumto that E[ x] 0 whch meas that OLS estmates of wll

More information

9 U-STATISTICS. Eh =(m!) 1 Eh(X (1),..., X (m ) ) i.i.d

9 U-STATISTICS. Eh =(m!) 1 Eh(X (1),..., X (m ) ) i.i.d 9 U-STATISTICS Suppose,,..., are P P..d. wth CDF F. Our goal s to estmate the expectato t (P)=Eh(,,..., m ). Note that ths expectato requres more tha oe cotrast to E, E, or Eh( ). Oe example s E or P((,

More information

The number of observed cases The number of parameters. ith case of the dichotomous dependent variable. the ith case of the jth parameter

The number of observed cases The number of parameters. ith case of the dichotomous dependent variable. the ith case of the jth parameter LOGISTIC REGRESSION Notato Model Logstc regresso regresses a dchotomous depedet varable o a set of depedet varables. Several methods are mplemeted for selectg the depedet varables. The followg otato s

More information

Training Sample Model: Given n observations, [[( Yi, x i the sample model can be expressed as (1) where, zero and variance σ

Training Sample Model: Given n observations, [[( Yi, x i the sample model can be expressed as (1) where, zero and variance σ Stat 74 Estmato for Geeral Lear Model Prof. Goel Broad Outle Geeral Lear Model (GLM): Trag Samle Model: Gve observatos, [[( Y, x ), x = ( x,, xr )], =,,, the samle model ca be exressed as Y = µ ( x, x,,

More information

ρ < 1 be five real numbers. The

ρ < 1 be five real numbers. The Lecture o BST 63: Statstcal Theory I Ku Zhag, /0/006 Revew for the prevous lecture Deftos: covarace, correlato Examples: How to calculate covarace ad correlato Theorems: propertes of correlato ad covarace

More information

Simulation Output Analysis

Simulation Output Analysis Smulato Output Aalyss Summary Examples Parameter Estmato Sample Mea ad Varace Pot ad Iterval Estmato ermatg ad o-ermatg Smulato Mea Square Errors Example: Sgle Server Queueg System x(t) S 4 S 4 S 3 S 5

More information

LINEAR REGRESSION ANALYSIS

LINEAR REGRESSION ANALYSIS LINEAR REGRESSION ANALYSIS MODULE V Lecture - Correctg Model Iadequaces Through Trasformato ad Weghtg Dr. Shalabh Departmet of Mathematcs ad Statstcs Ida Isttute of Techology Kapur Aalytcal methods for

More information

STATISTICAL INFERENCE

STATISTICAL INFERENCE (STATISTICS) STATISTICAL INFERENCE COMPLEMENTARY COURSE B.Sc. MATHEMATICS III SEMESTER ( Admsso) UNIVERSITY OF CALICUT SCHOOL OF DISTANCE EDUCATION CALICUT UNIVERSITY P.O., MALAPPURAM, KERALA, INDIA -

More information

Chapter 4 Multiple Random Variables

Chapter 4 Multiple Random Variables Revew o BST 63: Statstcal Theory I Ku Zhag, /0/008 Revew for Chapter 4-5 Notes: Although all deftos ad theorems troduced our lectures ad ths ote are mportat ad you should be famlar wth, but I put those

More information

CHAPTER VI Statistical Analysis of Experimental Data

CHAPTER VI Statistical Analysis of Experimental Data Chapter VI Statstcal Aalyss of Expermetal Data CHAPTER VI Statstcal Aalyss of Expermetal Data Measuremets do ot lead to a uque value. Ths s a result of the multtude of errors (maly radom errors) that ca

More information

Introduction to Matrices and Matrix Approach to Simple Linear Regression

Introduction to Matrices and Matrix Approach to Simple Linear Regression Itroducto to Matrces ad Matrx Approach to Smple Lear Regresso Matrces Defto: A matrx s a rectagular array of umbers or symbolc elemets I may applcatos, the rows of a matrx wll represet dvduals cases (people,

More information

Dr. Shalabh. Indian Institute of Technology Kanpur

Dr. Shalabh. Indian Institute of Technology Kanpur Aalyss of Varace ad Desg of Expermets-I MODULE -I LECTURE - SOME RESULTS ON LINEAR ALGEBRA, MATRIX THEORY AND DISTRIBUTIONS Dr. Shalabh Departmet t of Mathematcs t ad Statstcs t t Ida Isttute of Techology

More information

Bayes Estimator for Exponential Distribution with Extension of Jeffery Prior Information

Bayes Estimator for Exponential Distribution with Extension of Jeffery Prior Information Malaysa Joural of Mathematcal Sceces (): 97- (9) Bayes Estmator for Expoetal Dstrbuto wth Exteso of Jeffery Pror Iformato Hadeel Salm Al-Kutub ad Noor Akma Ibrahm Isttute for Mathematcal Research, Uverst

More information

Unsupervised Learning and Other Neural Networks

Unsupervised Learning and Other Neural Networks CSE 53 Soft Computg NOT PART OF THE FINAL Usupervsed Learg ad Other Neural Networs Itroducto Mture Destes ad Idetfablty ML Estmates Applcato to Normal Mtures Other Neural Networs Itroducto Prevously, all

More information

{ }{ ( )} (, ) = ( ) ( ) ( ) Chapter 14 Exercises in Sampling Theory. Exercise 1 (Simple random sampling): Solution:

{ }{ ( )} (, ) = ( ) ( ) ( ) Chapter 14 Exercises in Sampling Theory. Exercise 1 (Simple random sampling): Solution: Chapter 4 Exercses Samplg Theory Exercse (Smple radom samplg: Let there be two correlated radom varables X ad A sample of sze s draw from a populato by smple radom samplg wthout replacemet The observed

More information

Lecture 9: Tolerant Testing

Lecture 9: Tolerant Testing Lecture 9: Tolerat Testg Dael Kae Scrbe: Sakeerth Rao Aprl 4, 07 Abstract I ths lecture we prove a quas lear lower boud o the umber of samples eeded to do tolerat testg for L dstace. Tolerat Testg We have

More information

Lecture Notes 2. The ability to manipulate matrices is critical in economics.

Lecture Notes 2. The ability to manipulate matrices is critical in economics. Lecture Notes. Revew of Matrces he ablt to mapulate matrces s crtcal ecoomcs.. Matr a rectagular arra of umbers, parameters, or varables placed rows ad colums. Matrces are assocated wth lear equatos. lemets

More information

Chapter 13 Student Lecture Notes 13-1

Chapter 13 Student Lecture Notes 13-1 Chapter 3 Studet Lecture Notes 3- Basc Busess Statstcs (9 th Edto) Chapter 3 Smple Lear Regresso 4 Pretce-Hall, Ic. Chap 3- Chapter Topcs Types of Regresso Models Determg the Smple Lear Regresso Equato

More information

22 Nonparametric Methods.

22 Nonparametric Methods. 22 oparametrc Methods. I parametrc models oe assumes apror that the dstrbutos have a specfc form wth oe or more ukow parameters ad oe tres to fd the best or atleast reasoably effcet procedures that aswer

More information

ECON 482 / WH Hong The Simple Regression Model 1. Definition of the Simple Regression Model

ECON 482 / WH Hong The Simple Regression Model 1. Definition of the Simple Regression Model ECON 48 / WH Hog The Smple Regresso Model. Defto of the Smple Regresso Model Smple Regresso Model Expla varable y terms of varable x y = β + β x+ u y : depedet varable, explaed varable, respose varable,

More information

THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA

THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA THE ROYAL STATISTICAL SOCIETY EXAMINATIONS SOLUTIONS GRADUATE DIPLOMA PAPER II STATISTICAL THEORY & METHODS The Socety provdes these solutos to assst caddates preparg for the examatos future years ad for

More information

Multiple Linear Regression Analysis

Multiple Linear Regression Analysis LINEA EGESSION ANALYSIS MODULE III Lecture - 4 Multple Lear egresso Aalyss Dr. Shalabh Departmet of Mathematcs ad Statstcs Ida Isttute of Techology Kapur Cofdece terval estmato The cofdece tervals multple

More information

Investigation of Partially Conditional RP Model with Response Error. Ed Stanek

Investigation of Partially Conditional RP Model with Response Error. Ed Stanek Partally Codtoal Radom Permutato Model 7- vestgato of Partally Codtoal RP Model wth Respose Error TRODUCTO Ed Staek We explore the predctor that wll result a smple radom sample wth respose error whe a

More information

STA 108 Applied Linear Models: Regression Analysis Spring Solution for Homework #1

STA 108 Applied Linear Models: Regression Analysis Spring Solution for Homework #1 STA 08 Appled Lear Models: Regresso Aalyss Sprg 0 Soluto for Homework #. Let Y the dollar cost per year, X the umber of vsts per year. The the mathematcal relato betwee X ad Y s: Y 300 + X. Ths s a fuctoal

More information

Chapter 2 Supplemental Text Material

Chapter 2 Supplemental Text Material -. Models for the Data ad the t-test Chapter upplemetal Text Materal The model preseted the text, equato (-3) s more properl called a meas model. ce the mea s a locato parameter, ths tpe of model s also

More information

Point Estimation: definition of estimators

Point Estimation: definition of estimators Pot Estmato: defto of estmators Pot estmator: ay fucto W (X,..., X ) of a data sample. The exercse of pot estmato s to use partcular fuctos of the data order to estmate certa ukow populato parameters.

More information

Class 13,14 June 17, 19, 2015

Class 13,14 June 17, 19, 2015 Class 3,4 Jue 7, 9, 05 Pla for Class3,4:. Samplg dstrbuto of sample mea. The Cetral Lmt Theorem (CLT). Cofdece terval for ukow mea.. Samplg Dstrbuto for Sample mea. Methods used are based o CLT ( Cetral

More information

Discrete Choice Modeling

Discrete Choice Modeling Dscrete Choce Modelg Wllam Greee Ster School of Busess New York Uversty [Topc 6-Nolear Models] 1/87 6. NONLINEAR MODELS [Topc 6-Nolear Models] 2/87 Nolear Models Ageda Estmato Theory for Nolear Models

More information

Continuous Distributions

Continuous Distributions 7//3 Cotuous Dstrbutos Radom Varables of the Cotuous Type Desty Curve Percet Desty fucto, f (x) A smooth curve that ft the dstrbuto 3 4 5 6 7 8 9 Test scores Desty Curve Percet Probablty Desty Fucto, f

More information

Chapter 9 Jordan Block Matrices

Chapter 9 Jordan Block Matrices Chapter 9 Jorda Block atrces I ths chapter we wll solve the followg problem. Gve a lear operator T fd a bass R of F such that the matrx R (T) s as smple as possble. f course smple s a matter of taste.

More information

b. There appears to be a positive relationship between X and Y; that is, as X increases, so does Y.

b. There appears to be a positive relationship between X and Y; that is, as X increases, so does Y. .46. a. The frst varable (X) s the frst umber the par ad s plotted o the horzotal axs, whle the secod varable (Y) s the secod umber the par ad s plotted o the vertcal axs. The scatterplot s show the fgure

More information

DISTURBANCE TERMS. is a scalar and x i

DISTURBANCE TERMS. is a scalar and x i DISTURBANCE TERMS I a feld of research desg, we ofte have the qesto abot whether there s a relatoshp betwee a observed varable (sa, ) ad the other observed varables (sa, x ). To aswer the qesto, we ma

More information

Comparing Different Estimators of three Parameters for Transmuted Weibull Distribution

Comparing Different Estimators of three Parameters for Transmuted Weibull Distribution Global Joural of Pure ad Appled Mathematcs. ISSN 0973-768 Volume 3, Number 9 (207), pp. 55-528 Research Ida Publcatos http://www.rpublcato.com Comparg Dfferet Estmators of three Parameters for Trasmuted

More information

Bayes (Naïve or not) Classifiers: Generative Approach

Bayes (Naïve or not) Classifiers: Generative Approach Logstc regresso Bayes (Naïve or ot) Classfers: Geeratve Approach What do we mea by Geeratve approach: Lear p(y), p(x y) ad the apply bayes rule to compute p(y x) for makg predctos Ths s essetally makg

More information

Linear Regression Linear Regression with Shrinkage. Some slides are due to Tommi Jaakkola, MIT AI Lab

Linear Regression Linear Regression with Shrinkage. Some slides are due to Tommi Jaakkola, MIT AI Lab Lear Regresso Lear Regresso th Shrkage Some sldes are due to Tomm Jaakkola, MIT AI Lab Itroducto The goal of regresso s to make quattatve real valued predctos o the bass of a vector of features or attrbutes.

More information

Mean is only appropriate for interval or ratio scales, not ordinal or nominal.

Mean is only appropriate for interval or ratio scales, not ordinal or nominal. Mea Same as ordary average Sum all the data values ad dvde by the sample sze. x = ( x + x +... + x Usg summato otato, we wrte ths as x = x = x = = ) x Mea s oly approprate for terval or rato scales, ot

More information

1 Solution to Problem 6.40

1 Solution to Problem 6.40 1 Soluto to Problem 6.40 (a We wll wrte T τ (X 1,...,X where the X s are..d. wth PDF f(x µ, σ 1 ( x µ σ g, σ where the locato parameter µ s ay real umber ad the scale parameter σ s > 0. Lettg Z X µ σ we

More information

Comparison of Dual to Ratio-Cum-Product Estimators of Population Mean

Comparison of Dual to Ratio-Cum-Product Estimators of Population Mean Research Joural of Mathematcal ad Statstcal Sceces ISS 30 6047 Vol. 1(), 5-1, ovember (013) Res. J. Mathematcal ad Statstcal Sc. Comparso of Dual to Rato-Cum-Product Estmators of Populato Mea Abstract

More information

Midterm Exam 1, section 2 (Solution) Thursday, February hour, 15 minutes

Midterm Exam 1, section 2 (Solution) Thursday, February hour, 15 minutes coometrcs, CON Sa Fracsco State Uverst Mchael Bar Sprg 5 Mdterm xam, secto Soluto Thursda, Februar 6 hour, 5 mutes Name: Istructos. Ths s closed book, closed otes exam.. No calculators of a kd are allowed..

More information

Midterm Exam 1, section 1 (Solution) Thursday, February hour, 15 minutes

Midterm Exam 1, section 1 (Solution) Thursday, February hour, 15 minutes coometrcs, CON Sa Fracsco State Uversty Mchael Bar Sprg 5 Mdterm am, secto Soluto Thursday, February 6 hour, 5 mutes Name: Istructos. Ths s closed book, closed otes eam.. No calculators of ay kd are allowed..

More information

Discrete Mathematics and Probability Theory Fall 2016 Seshia and Walrand DIS 10b

Discrete Mathematics and Probability Theory Fall 2016 Seshia and Walrand DIS 10b CS 70 Dscrete Mathematcs ad Probablty Theory Fall 206 Sesha ad Walrad DIS 0b. Wll I Get My Package? Seaky delvery guy of some compay s out delverg packages to customers. Not oly does he had a radom package

More information

18.413: Error Correcting Codes Lab March 2, Lecture 8

18.413: Error Correcting Codes Lab March 2, Lecture 8 18.413: Error Correctg Codes Lab March 2, 2004 Lecturer: Dael A. Spelma Lecture 8 8.1 Vector Spaces A set C {0, 1} s a vector space f for x all C ad y C, x + y C, where we take addto to be compoet wse

More information

Estimation of Stress- Strength Reliability model using finite mixture of exponential distributions

Estimation of Stress- Strength Reliability model using finite mixture of exponential distributions Iteratoal Joural of Computatoal Egeerg Research Vol, 0 Issue, Estmato of Stress- Stregth Relablty model usg fte mxture of expoetal dstrbutos K.Sadhya, T.S.Umamaheswar Departmet of Mathematcs, Lal Bhadur

More information

Dimensionality Reduction and Learning

Dimensionality Reduction and Learning CMSC 35900 (Sprg 009) Large Scale Learg Lecture: 3 Dmesoalty Reducto ad Learg Istructors: Sham Kakade ad Greg Shakharovch L Supervsed Methods ad Dmesoalty Reducto The theme of these two lectures s that

More information

Lecture Notes Forecasting the process of estimating or predicting unknown situations

Lecture Notes Forecasting the process of estimating or predicting unknown situations Lecture Notes. Ecoomc Forecastg. Forecastg the process of estmatg or predctg ukow stuatos Eample usuall ecoomsts predct future ecoomc varables Forecastg apples to a varet of data () tme seres data predctg

More information

LECTURE - 4 SIMPLE RANDOM SAMPLING DR. SHALABH DEPARTMENT OF MATHEMATICS AND STATISTICS INDIAN INSTITUTE OF TECHNOLOGY KANPUR

LECTURE - 4 SIMPLE RANDOM SAMPLING DR. SHALABH DEPARTMENT OF MATHEMATICS AND STATISTICS INDIAN INSTITUTE OF TECHNOLOGY KANPUR amplg Theory MODULE II LECTURE - 4 IMPLE RADOM AMPLIG DR. HALABH DEPARTMET OF MATHEMATIC AD TATITIC IDIA ITITUTE OF TECHOLOGY KAPUR Estmato of populato mea ad populato varace Oe of the ma objectves after

More information

Example: Multiple linear regression. Least squares regression. Repetition: Simple linear regression. Tron Anders Moger

Example: Multiple linear regression. Least squares regression. Repetition: Simple linear regression. Tron Anders Moger Example: Multple lear regresso 5000,00 4000,00 Tro Aders Moger 0.0.007 brthweght 3000,00 000,00 000,00 0,00 50,00 00,00 50,00 00,00 50,00 weght pouds Repetto: Smple lear regresso We defe a model Y = β0

More information

Lecture 8: Linear Regression

Lecture 8: Linear Regression Lecture 8: Lear egresso May 4, GENOME 56, Sprg Goals Develop basc cocepts of lear regresso from a probablstc framework Estmatg parameters ad hypothess testg wth lear models Lear regresso Su I Lee, CSE

More information

THE ROYAL STATISTICAL SOCIETY HIGHER CERTIFICATE

THE ROYAL STATISTICAL SOCIETY HIGHER CERTIFICATE THE ROYAL STATISTICAL SOCIETY 00 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE PAPER I STATISTICAL THEORY The Socety provdes these solutos to assst caddates preparg for the examatos future years ad for the

More information

Regresso What s a Model? 1. Ofte Descrbe Relatoshp betwee Varables 2. Types - Determstc Models (o radomess) - Probablstc Models (wth radomess) EPI 809/Sprg 2008 9 Determstc Models 1. Hypothesze

More information

Handout #8. X\Y f(x) 0 1/16 1/ / /16 3/ / /16 3/16 0 3/ /16 1/16 1/8 g(y) 1/16 1/4 3/8 1/4 1/16 1

Handout #8. X\Y f(x) 0 1/16 1/ / /16 3/ / /16 3/16 0 3/ /16 1/16 1/8 g(y) 1/16 1/4 3/8 1/4 1/16 1 Hadout #8 Ttle: Foudatos of Ecoometrcs Course: Eco 367 Fall/05 Istructor: Dr. I-Mg Chu Lear Regresso Model So far we have focused mostly o the study of a sgle radom varable, ts correspodg theoretcal dstrbuto,

More information