Multi-Equation Structural Models: Seemingly Unrelated Regression Models

Size: px
Start display at page:

Download "Multi-Equation Structural Models: Seemingly Unrelated Regression Models"

Transcription

1 Chapter 15 Multi-Equation Structural Models: Seemingly Unrelated Regression Models

2 Section 15.1 Seemingly Unrelated Regression Models

3 Modeling Approaches Econometric (Structural) Models Time-Series Models Intuitive Models Single- Equation Multi- Equation Univariate Multivariate Univariate Simultaneous Models Recursive Models Seemingly Unrelated Regression Models Box- Jenkins ARIMA Models Error Correction Models Vector Autoregression (VAR) Models Seasonal Smoothing Models Models Trend Extrapolation Models 3 In some instances, econometric models might encompass elements of timeseries models (for example, ARCH and GARCH models, adjustments for serial correlation).

4 Seemingly Unrelated Regression Models (SUR) Y Y : Y 1t 2t mt = = = β β 11 β 21 X m1 X 1t,1 X 2t,1 mt,1 + β 12 + β 22 + β X m2 1t,2 X 2t,2 X β mt,2 1k β X 2k β 1t, k1 X mkm + ε 2t, k 2 X 1t + ε mt, km 2t + ε SUR parameter estimates are full-information coefficients, whereas OLS parameter estimates are limited-information coefficients. mt Σ = σ σ m σ... σ 1m mm diagonal elements are variances; off-diagonal elements are covariances. 4 (1) If the RHS variables are the same across the m equations, the SUR parameter estimates are the same as the OLS parameter estimates even if Σ is not a diagonal matrix. (2) If the RHS variables are not the same across the m equations and if Σ is a diagonal matrix, then the SUR parameter estimates are the same as the OLS parameter estimates.

5 Section 15.2 Example of Seemingly Unrelated Regression Models

6 Example of Seemingly Unrelated Regression Models (Zellner 1962) I tm β β β + ε = 0m + 1mCt 1 + 2mFt 1 mt I = Gross Investment C = End-of-Period Capital Stock F = End-of-Period Value of Outstanding Shares Annual Data for Two Firms General Electric (GE) and Westinghouse (W) continued...

7 Σ is not a diagonal matrix represents the correlation of the disturbance terms in the GE and W equations. OLS SUR GE I W I t GE I W I t t = ( ) = t (8.0153) = ( ) = (7.5452) C t 1 (0.0257) C t 1 (0.0561) C C t 1 (0.0250) (0.053) t F t 1 (0.0156) F t 1 (0.0157) F F t 1 (0.0145) t 1 (0.0145) (standard errors in parentheses) 7 The use of SUR results in lower standard errors vis-àvis the use of OLS.

8 Section 15.3 Example: Demand for a Cereal Product from Five Retailers: HEB, Publix, Food Lion, Fred Myer, and Meijer

9 Example: Demand for a Cereal Product from Five Retailers: HEB, Publix, Food Lion, Fred Myer, and Meijer (No Adjustment for Serial Correlation) 9 SAS Program for Proc Syslin Use of Seemingly Unrelated Regression data HEB; input heb_retailer product week heb_logdisc heb_logprice heb_fsi1 weekno heb_logunits heb_logdisp heb_logad heb_logdist q1 q2 q3 q4; datalines; run; data PUBLIX; input publix_retailer product week publix_logdisc publix_logprice publix_fsi1 weekno publix_logunits publix_logdisp publix_logad publix_logdist q1 q2 q3 q4; datalines; run; data FOODLION; input flion_retailer product week flion_logdisc flion_logprice flion_fsi1 weekno flion_logunits flion_logdisp flion_logad flion_logdist q1 q2 q3 q4; datalines; run;

10 data FREDMYER; input fmyer_retailer product week fmyer_logdisc fmyer_logprice fmyer_fsi1 weekno fmyer_logunits fmyer_logdisp fmyer_logad fmyer_logdist q1 q2 q3 q4; datalines; run; data MEIJER; input meijer_retailer product week meijer_logdisc meijer_logprice meijer_fsi1 weekno meijer_logunits meijer_logdisp meijer_logad meijer_logdist q1 q2 q3 q4; datalines; run; options nodate; data all; merge heb publix foodlion fredmyer meijer; by week; proc means data=all n mean median std min max; var heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits; run; proc means data=all n mean median std min max; var heb_logprice publix_logprice flion_logprice fmyer_logprice meijer_logprice; run; 10

11 proc syslin data=all sur out=retailersur; heb: model heb_logunits=heb_logprice heb_logdisc heb_fsi1 heb_logdisp heb_logad heb_logdist weekno q1 q2 q3 / dw; output predicted=pheb_logunits; publix: model publix_logunits=publix_logprice publix_logdisc publix_fsi1 publix_logdisp publix_logad publix_logdist weekno q1 q2 q3 / dw; output predicted=ppublix_logunits; flion: model flion_logunits=flion_logprice flion_logdisc flion_fsi1 flion_logdisp flion_logad flion_logdist weekno q1 q2 q3 / dw; output predicted=pflion_logunits; fmyer: model fmyer_logunits=fmyer_logprice fmyer_logdisc fmyer_fsi1 fmyer_logdisp fmyer_logad fmyer_logdist weekno q1 q2 q3 / dw; output predicted=pfmyer_logunits; meijer: model meijer_logunits=meijer_logprice meijer_logdisc meijer_fsi1 meijer_logdisp meijer_logad meijer_logdist weekno q1 q2 q3 / dw; output predicted=pmeijer_logunits; 11

12 stest heb.heb_logprice-publix.publix_logprice=0; stest heb.heb_logprice-flion.flion_logprice=0; stest heb.heb_logprice-fmyer.fmyer_logprice=0; stest heb.heb_logprice-meijer.meijer_logprice=0; stest publix.publix_logprice-flion.flion_logprice=0; stest publix.publix_logprice-fmyer.fmyer_logprice=0; stest publix.publix_logprice-meijer.meijer_logprice=0; stest flion.flion_logprice-fmyer.fmyer_logprice=0; stest flion.flion_logprice-meijer.meijer_logprice=0; stest fmyer.fmyer_logprice-meijer.meijer_logprice=0; run; System tests of the own-price elasticities 12

13 proc corr data=retailersur; var heb_logunits pheb_logunits; run; proc corr data=retailersur; var publix_logunits ppublix_logunits; run; proc corr data=retailersur; var flion_logunits pflion_logunits; run; proc corr data=retailersur; var fmyer_logunits pfmyer_logunits; run; proc corr data=retailersur; var meijer_logunits pmeijer_logunits; run; 13

14 The MEANS Procedure Variable N Mean Median Std Dev Minimum Maximum ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ The MEANS Procedure Variable N Mean Median Std Dev Minimum Maximum ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ heb_logprice publix_logprice flion_logprice fmyer_logprice meijer_logprice ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ 14

15 The SYSLIN Procedure Ordinary Least Squares Estimation Model HEB Dependent Variable heb_logunits Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

16 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 heb_logprice heb_logdisc heb_fsi heb_logdisp <.0001 heb_logad heb_logdist <.0001 weekno q q q

17 The SYSLIN Procedure Ordinary Least Squares Estimation Model PUBLIX Dependent Variable publix_logunits Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

18 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 publix_logprice <.0001 publix_logdisc <.0001 publix_fsi publix_logdisp <.0001 publix_logad <.0001 publix_logdist weekno q q q

19 The SYSLIN Procedure Ordinary Least Squares Estimation Model FLION Dependent Variable flion_logunits Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

20 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept flion_logprice flion_logdisc <.0001 flion_fsi flion_logdisp <.0001 flion_logad flion_logdist <.0001 weekno q q q

21 The SYSLIN Procedure Ordinary Least Squares Estimation Model FMYER Dependent Variable fmyer_logunits Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

22 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 fmyer_logprice <.0001 fmyer_logdisc <.0001 fmyer_fsi fmyer_logdisp <.0001 fmyer_logad fmyer_logdist weekno q q q

23 The SYSLIN Procedure Ordinary Least Squares Estimation Model MEIJER Dependent Variable meijer_logunits Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

24 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept meijer_logprice meijer_logdisc <.0001 meijer_fsi meijer_logdisp <.0001 meijer_logad meijer_logdist weekno q q q

25 The SYSLIN Procedure Seemingly Unrelated Regression Estimation Cross Model Covariance HEB PUBLIX FLION FMYER MEIJER HEB PUBLIX FLION FMYER MEIJER Cross Model Correlation Σ matrix in correlation form HEB PUBLIX FLION FMYER MEIJER 25 HEB PUBLIX FLION FMYER MEIJER

26 Cross Model Inverse Correlation HEB PUBLIX FLION FMYER MEIJER HEB PUBLIX FLION FMYER MEIJER Cross Model Inverse Covariance HEB PUBLIX FLION FMYER MEIJER HEB PUBLIX FLION FMYER MEIJER System Weighted MSE Degrees of freedom 750 System Weighted R-Square

27 The SYSLIN Procedure Seemingly Unrelated Regression Estimation Model HEB Dependent Variable heb_logunits Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t 27 Intercept <.0001 heb_logprice heb_logdisc heb_fsi heb_logdisp <.0001 heb_logad heb_logdist <.0001 weekno q q q

28 Durbin-Watson Number of Observations 161 First-Order Autocorrelation Model PUBLIX Dependent Variable publix_logunits Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 publix_logprice <.0001 publix_logdisc <.0001 publix_fsi publix_logdisp <.0001 publix_logad <.0001 publix_logdist weekno q q q

29 The SYSLIN Procedure Seemingly Unrelated Regression Estimation Durbin-Watson Number of Observations 161 First-Order Autocorrelation Model FLION Dependent Variable flion_logunits Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t 29 Intercept flion_logprice flion_logdisc <.0001 flion_fsi flion_logdisp <.0001 flion_logad flion_logdist <.0001 weekno q q q

30 Durbin-Watson Number of Observations 161 First-Order Autocorrelation Model FMYER Dependent Variable fmyer_logunits Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 fmyer_logprice <.0001 fmyer_logdisc <.0001 fmyer_fsi fmyer_logdisp <.0001 fmyer_logad fmyer_logdist weekno q q q

31 Durbin-Watson Number of Observations 161 First-Order Autocorrelation Model MEIJER Dependent Variable meijer_logunits 31 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 meijer_logprice meijer_logdisc <.0001 meijer_fsi meijer_logdisp <.0001 meijer_logad meijer_logdist weekno q q q

32 Durbin-Watson Number of Observations 161 First-Order Autocorrelation Test Results of Hypotheses Regarding Own-Price Elasticites Num DF Den DF F Value Pr > F (1) H 0 : HEB = PUBLIX (2) (3) H 0 : HEB = FOOD LION H 0 : HEB = FRED MYER (4) H 0 : HEB = MEIJER (5) (6) H 0 : PUBLIX = FOOD LION H 0 : PUBLIX = FRED MYER 32

33 Num DF Den DF F Value Pr > F (7) H 0 : PUBLIX = MEIJER (8) H 0 : FOOD LION = FRED MYER (9) H 0 : FOOD LION = MEIJER (10) H 0 : FRED MYER = MEIJER 33

34 The CORR Procedure 2 Variables: heb_logunits pheb_logunits Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum heb_logunits pheb_logunits Simple Statistics Variable Label heb_logunits pheb_logunits Predicted Values 34

35 Pearson Correlation Coefficients, N = 164 Prob > r under H0: Rho=0 heb_ pheb_ logunits logunits heb_logunits <.0001 pheb_logunits Predicted Values <.0001 R 2 = (.83765) 2 = The CORR Procedure 2 Variables: publix_logunits ppublix_logunits Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum publix_logunits ppublix_logunits

36 Simple Statistics Variable Label publix_logunits ppublix_logunits Predicted Values Pearson Correlation Coefficients, N = 165 Prob > r under H0: Rho=0 publix_ ppublix_ logunits logunits publix_logunits <.0001 ppublix_logunits Predicted Values < R 2 = (.97490) 2 =

37 The CORR Procedure 2 Variables: flion_logunits pflion_logunits Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum flion_logunits pflion_logunits Simple Statistics Variable Label flion_logunits pflion_logunits Predicted Values 37

38 Pearson Correlation Coefficients, N = 165 Prob > r under H0: Rho=0 flion_ pflion_ logunits logunits flion_logunits <.0001 pflion_logunits Predicted Values <.0001 R 2 = (.98733) 2 =

39 The CORR Procedure 2 Variables: fmyer_logunits pfmyer_logunits Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum fmyer_logunits pfmyer_logunits Simple Statistics Variable Label fmyer_logunits pfmyer_logunits Predicted Values 39

40 Pearson Correlation Coefficients, N = 165 Prob > r under H0: Rho=0 fmyer_ pfmyer_ logunits logunits fmyer_logunits <.0001 pfmyer_logunits Predicted Values <.0001 R 2 = (.94580) 2 =

41 The CORR Procedure 2 Variables: meijer_logunits pmeijer_logunits Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum meijer_logunits pmeijer_logunits Simple Statistics Variable Label meijer_logunits pmeijer_logunits Predicted Values 41

42 Pearson Correlation Coefficients, N = 162 Prob > r under H0: Rho=0 meijer_ pmeijer_ logunits logunits meijer_logunits <.0001 pmeijer_logunits Predicted Values <.0001 R 2 = (.96525) 2 =

43 Use of Proc Model: SUR Analysis With and Without Adjustment for Serial Correlation options nodate; data all; merge heb publix foodlion fredmyer meijer; by week; proc means data=all n mean median std min max; var heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits; run; proc means data=all n mean median std min max; var heb_logprice publix_logprice flion_logprice fmyer_logprice meijer_logprice; run; * without adjustment for serial correlation; 43

44 proc model data=all; heb_logunits=a0+a1*heb_logprice+a2*heb_logdisc+a3*heb_fsi1+ a4*heb_logdisp+a5*heb_logad+a6*heb_logdist+ a7*weekno+a8*q1+a9*q2+a10*q3; publix_logunits=b0+b1*publix_logprice+b2*publix_logdisc+ b3*publix_fsi1+b4*publix_logdisp+ b5*publix_logad+b6*publix_logdist+b7*weekno+b8*q1+b9*q2+b10*q3; flion_logunits=c0+c1*flion_logprice+c2*flion_logdisc+c3*flion_fsi1+ c4*flion_logdisp+c5*flion_logad+ c6*flion_logdist+c7*weekno+c8*q1+c9*q2+c10*q3; fmyer_logunits=d0+d1*fmyer_logprice+d2*fmyer_logdisc+d3*fmyer_fsi1+ d4*fmyer_logdisp+d5*fmyer_logad+ d6*fmyer_logdist+d7*weekno+d8*q1+d9*q2+d10*q3; meijer_logunits=e0+e1*meijer_logprice+e2*meijer_logdisc+ e3*meijer_fsi1+e4*meijer_logdisp+e5*meijer_logad+ e6*meijer_logdist+e7*weekno+e8*q1+e9*q2+e10*q3; 44

45 fit heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits / sur dw dwprob normal out=retailersur; parms a0 a1 a2 a3 a4 a5 a6 a7 a8 a9 a10 b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 c10 d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10 e0 e1 e2 e3 e4 e5 e6 e7 e8 e9 e10; test a1-b1=0; test a1-c1=0; test a1-d1=0; test a1-e1=0; test b1-c1=0; test b1-d1=0; test b1-e1=0; test c1-d1=0; test c1-e1=0; test d1-e1=0; run; * with adjustment for serial correlation; 45

46 46 proc model data=all; heb_logunits=a0+a1*heb_logprice+a2*heb_logdisc+a3* heb_fsi1+a4*heb_logdisp+a5*heb_logad+a6*heb_logdist+ a7*weekno+a8*q1+a9*q2+a10*q3; publix_logunits=b0+b1*publix_logprice+b2*publix_logdisc+ b3*publix_fsi1+b4*publix_logdisp+ b5*publix_logad+b6*publix_logdist+b7*weekno+b8*q1+ b9*q2+b10*q3; flion_logunits=c0+c1*flion_logprice+c2*flion_logdisc+ c3*flion_fsi1+c4*flion_logdisp+c5*flion_logad+ c6*flion_logdist+c7*weekno+c8*q1+c9*q2+c10*q3; fmyer_logunits=d0+d1*fmyer_logprice+d2*fmyer_logdisc+ d3*fmyer_fsi1+d4*fmyer_logdisp+d5*fmyer_logad+ d6*fmyer_logdist+d7*weekno+d8*q1+d9*q2+d10*q3; meijer_logunits=e0+e1*meijer_logprice+e2*meijer_logdisc+ e3*meijer_fsi1+e4*meijer_logdisp+e5*meijer_logad+ e6*meijer_logdist+e7*weekno+e8*q1+e9*q2+e10*q3; %ar(heb_logunits,1); %ar(publix_logunits,1); %ar(flion_logunits,1); %ar(fmyer_logunits,1); %ar(meijer_logunits,1); AR(1) process in each of the five equations which comprise the system of seemingly unrelated regressions

47 fit heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits / sur dw dwprob normal out=retailersur; parms a0 a1 a2 a3 a4 a5 a6 a7 a8 a9 a10 b0 b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 c10 d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10 e0 e1 e2 e3 e4 e5 e6 e7 e8 e9 e10; test a1-b1=0; test a1-c1=0; test a1-d1=0; test a1-e1=0; test b1-c1=0; test b1-d1=0; test b1-e1=0; test c1-d1=0; test c1-e1=0; test d1-e1=0; run; 47

48 Use of PROC Model to Obtain Estimates of Parameters from a System of Seemingly Unrelated Regressions The MODEL Procedure Model Summary Model Variables 5 Parameters 55 Equations 5 Number of Statements 25 The 5 Equations to Estimate 48

49 heb_logunits = F(a0(1), a1(heb_logprice), a2(heb_logdisc), a3(heb_fsi1), a4(heb_logdisp), a5(heb_logad), a6(heb_logdist), a7(weekno), a8(q1), a9(q2), a10(q3)) publix_logunits = F(b0(1), b1(publix_logprice), b2(publix_logdisc), b3(publix_fsi1), b4(publix_logdisp), b5(publix_logad), b6(publix_logdist), b7(weekno), b8(q1), b9(q2), b10(q3)) flion_logunits = F(c0(1), c1(flion_logprice), c2(flion_logdisc), c3(flion_fsi1), c4(flion_logdisp), c5(flion_logad), c6(flion_logdist), c7(weekno), c8(q1), c9(q2), c10(q3)) fmyer_logunits = F(d0(1), d1(fmyer_logprice), d2(fmyer_logdisc), d3(fmyer_fsi1), d4(fmyer_logdisp), d5(fmyer_logad), d6(fmyer_logdist), d7(weekno), d8(q1), d9(q2), d10(q3)) meijer_logunits = F(e0(1), e1(meijer_logprice), e2(meijer_logdisc), e3(meijer_fsi1), e4(meijer_logdisp), e5(meijer_logad), e6(meijer_logdist), e7(weekno), e8(q1), e9(q2), e10(q3)) 49 NOTE: At SUR Iteration 1 CONVERGE=0.001 Criteria Met.

50 The MODEL Procedure SUR Estimation Summary Data Set Options DATA= ALL OUT= RETAILERSUR Minimization Summary Parameters Estimated 55 Method Gauss Iterations 1 Final Convergence Criteria R 0 PPC 5.64E-11 RPC(e7) Object Trace(S) Objective Value Observations Processed 50 Read 165 Solved 165 Used 161 Missing 4

51 Number of parameters to estimate Difference between number of observations and number of parameters to estimate The MODEL Procedure Nonlinear SUR Summary of Residual Errors DF DF Adj Durbin Equation Model Error SSE MSE Root MSE R-Square R-Sq Watson heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits parameters to estimate, including the intercept 51

52 Nonlinear SUR Parameter Estimates Approx Approx Parameter Estimate Std Err t Value Pr > t 52 a <.0001 a a a a <.0001 a a <.0001 a a a a b <.0001 b <.0001 b <.0001 b b <.0001 b <.0001 b b7 Precisely the same estimates and standard b errors obtained with Proc SYSLIN b b

53 c c c <.0001 c c <.0001 c c <.0001 c c c c d <

54 d <.0001 d <.0001 d d <.0001 d d d d d d e <.0001 e e <.0001 e e <.0001 e e e e e e

55 Test Results SUR results no adjustment for serial correlation Test Type Statistic Pr > ChiSq Label Test0 Wald <.0001 a1-b1=0 Test1 Wald a1-c1=0 Test2 Wald <.0001 a1-d1=0 Test3 Wald a1-e1=0 Test4 Wald b1-c1=0 Test5 Wald b1-d1=0 Test6 Wald b1-e1=0 Test7 Wald c1-d1=0 Test8 Wald c1-e1=0 Test9 Wald d1-e1=0 Rejection of hypotheses Number of Observations Statistics for System 55 Used 161 Objective Missing 4 Objective*N

56 The MODEL Procedure Durbin-Watson Statistics Equation Order DW Pr < DW Pr > DW heb_logunits < publix_logunits < flion_logunits fmyer_logunits < meijer_logunits Evidence of AR(1) process in all equations. 56

57 Evidence concerning normality of residuals for individual equations and for the system as a whole except for the food lion equation; normality of residuals is evident in heb, publix, fred myer, and meijer equations. Normality Test Equation Test Statistic Value Prob heb_logunits Shapiro-Wilk W publix_logunits Shapiro-Wilk W flion_logunits Shapiro-Wilk W fmyer_logunits Shapiro-Wilk W meijer_logunits Shapiro-Wilk W System Mardia Skewness <.0001 Mardia Kurtosis 9.57 <.0001 Henze-Zirkler T For the system as a whole, normality of residuals is evident.

58 The NORMAL option in the FIT statement of Proc Model allows the construction of multivariate and univariate tests of normality of residuals. Shipiro-Wilk W test: a univariate test, one for each equation. The W test is performed because the sample size is less than 2,000 observations. Computation of the Kolmogorov-Smirnoff test statistic (the other univariate test associated with Proc Model) requires at least 2,000 observations. Mandia Skewness (1980): a test of multivariate skewness 2 (asymptotically distributed as χ ). Mandia Kurtosis: a test of multivariate Kurtosis (asymptotically distributed as normal). Henze-Zirkler T-statistic (1990): a test of multivariate normality (asymptotically distributed as log normal). 58

59 The MODEL Procedure SUR Estimation Summary Data Set Options DATA= ALL OUT= RETAILERSUR Minimization Summary Parameters Estimated 60 Method Gauss Iterations 4 Final Convergence Criteria R PPC(e3) RPC(e3) Object 2.727E-6 Trace(S) Objective Value Observations Processed 59 Read 165 Solved 165 Used 161 Missing 4

60 The MODEL Procedure Nonlinear SUR Summary of Residual Errors DF DF Adj Durbin Equation Model Error SSE MSE Root MSE R-Square R-Sq Watson heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits * With adjustment for serial correlation 60

61 Nonlinear SUR Parameter Estimates Approx Approx Parameter Estimate Std Err t Value Pr > t Label 61 a <.0001 a a a a <.0001 a a a a a a b <.0001 b b <.0001 b b <.0001 b <.0001 b b b b b

62 62 c c c <.0001 c c <.0001 c c <.0001 c c c c d <.0001 d <.0001 d <.0001 d d <.0001 d d d d d d

63 e e e <.0001 e e <.0001 e e e e e e heb_logunits_l <.0001 AR(heb_logunits) heb_logunits lag1 parameter publix_logunits_l <.0001 AR(publix_logunits) publix_logunits lag1 parameter flion_logunits_l AR(flion_logunits) flion_logunits lag1 parameter fmyer_logunits_l <.0001 AR(fmyer_logunits) fmyer_logunits lag1 parameter meijer_logunits_l AR(meijer_logunits) meijer_logunits lag1 parameter 63

64 Test Results (with adjustment for serial correlation) Test Type Statistic Pr > ChiSq Label Test0 Wald a1-b1=0 Test1 Wald a1-c1=0 Test2 Wald a1-d1=0 Test3 Wald a1-e1=0 Test4 Wald b1-c1=0 Test5 Wald b1-d1=0 Test6 Wald b1-e1=0 Test7 Wald c1-d1=0 Test8 Wald c1-e1=0 Test9 Wald d1-e1=0 Number of Observations Statistics for System 64 Used 161 Objective Missing 4 Objective*N

65 The MODEL Procedure Durbin-Watson Statistics Equation Order DW Pr < DW Pr > DW heb_logunits publix_logunits flion_logunits fmyer_logunits meijer_logunits Serial correlation no longer a problem in any equation. 65

66 Normality Test Equation Test Statistic Value Prob heb_logunits Shapiro-Wilk W publix_logunits Shapiro-Wilk W flion_logunits Shapiro-Wilk W fmyer_logunits Shapiro-Wilk W 0.95 <.0001 meijer_logunits Shapiro-Wilk W System Mardia Skewness <.0001 Mardia Kurtosis <.0001 Henze-Zirkler T Non-normality of residuals in heb, flion, and fmyer equations. Non-normality of residuals as a system. 66

67 Variable SUR Results Demand for a Cereal Product : H-E-B Parameter Estimate OLS SUR a SUR b Standard Error Parameter Estimate Standard Error Parameter Estimate Standard Error Intercept log price log DISC Fsi log DISP log AD log DIST week no q q q R c (0.7086) d e AR(1)

68 SUR Results Demand for a Cereal Product : PUBLIX Variable Parameter Estimate OLS SUR a SUR b Standard Error Parameter Estimate Standard Error Parameter Estimate Standard Error Intercept log price log DISC Fsi log DISP log AD log DIST week no q q q R c d e AR(1)

69 Variable SUR Results Demand for a Cereal Product : Food OLS SUR Lion a SUR b Parameter Estimate Standard Error Parameter Estimate Standard Error Parameter Estimate Standard Error Intercept log price log DISC Fsi log DISP log AD log DIST week no q q q R c (0.9752) d e AR(1)

70 SUR Results Demand for a Cereal Product : FRED MYER Variable Parameter Estimate OLS SUR a SUR b Standard Error Parameter Estimate Standard Error Parameter Estimate Standard Error Intercept log price log DISC Fsi log DISP log AD log DIST week no q q q R c d e AR(1)

71 SUR Results Demand for a Cereal Product : MEIJER Variable Parameter Estimate OLS SUR a SUR b Standard Error Parameter Estimate Standard Error Parameter Estimate Standard Error Intercept log price log DISC Fsi log DISP log AD log DIST week no Q Q q R c d e AR(1)

72 Section 15.4 Seemingly Unrelated Regression Models with Restrictions

73 Often models require restrictions to conform to theoretical developments. Consumer demand functions derived from neoclassical theory possess the mathematical properties of homogeneity, adding-up, symmetry, and negativity. When estimating demand models, economists often impose homogeneity and symmetry as parameter restrictions. Statistical rejection of theoretical restrictions is NOT unusual; however, it is commonplace to see theoretical restrictions imposed in spite of their statistical failure. 73

74 Section 15.5 Rotterdam Model

75 Rotterdam Model (Barten, 1965) 75 w * it Dq Dq it it where : Dp Dy w w y * it it t t jt = = = = = = = b i Dy [ q / q ] log it it p i log log it t k [ p / p ] [ y / y ] ( w q p it it t it / q it jt + y t t 1 w jt 1 it 1 ) w * kt Dp kt + j c ij Dp jt + ε RESTRICTIONS c ij = 0 j c = c = total expenditure at time period t. ij ji it (HOMOGENEITY) (SYMMETRY) This model involves logarithmic differences of quantities, prices, and total expenditure.

76 Section 15.6 Linear Approximate Almost Ideal Demand System (LA/AIDS) Model

77 Linear Approximate (LA) Almost Ideal Demand System (AIDS) (Deaton and Muellbauer, 1980) w it = α + i j γ ij log p jt + β log( y i t / P t ) + ε it log P t = k w kt log p kt (Stone Price Index) RESTRICTIONS j γ ij = 0 (HOMOGENEITY) 77 γ ij = γ ji (SYMMETRY) A very popular model among demand economists.

78 Section 15.7 Example: Demand Interrelationships for Spaghetti Sauces: LA/AIDS Model

79 Example: Demand Interrelationships for Spaghetti Sauces: Use of LA/AIDS Model with Restrictions Use of PROC SYSLIN * Demonstration of Seemingly Unrelated Regression Model with Restrictions; * Demand Analysis of Various Spaghetti Sauces; data spaghettisauce; input WEEK QCLASSICO PCLASSICO QHUNTS PHUNTS QNEWMAN PNEWMAN QPREGO PPREGO QPRIVLABEL PPRIVLABEL QRAGU PRAGU TOTALEXP; wclassico=qclassico*pclassico/totalexp; whunts=qhunts*phunts/totalexp; wnewman=qnewman*pnewman/totalexp; wprego=qprego*pprego/totalexp; wprivlabel=qprivlabel*pprivlabel/totalexp; wragu=qragu*pragu/totalexp; 79

80 wcheck=wclassico+whunts+wnewman+wprego+ wprivlabel+wragu; lpclassico=log(pclassico); lphunts=log(phunts); lpnewman=log(pnewman); lpprego=log(pprego); lpprivlabel=log(pprivlabel); lpragu=log(pragu); StonePI=wclassico*lpclassico+whunts*lphunts+ wnewman*lpnewman+wprego*lpprego+ wprivlabel*lpprivlabel+wragu*lpragu; ltotexp=log(totalexp); lrealexp=ltotexp-stonepi; datalines; 80

81 options nodate; proc means data=spaghettisauce n mean median std min max; var QCLASSICO QHUNTS QNEWMAN QPREGO QPRIVLABEL QRAGU; run; proc means data=spaghettisauce n mean median std min max; var PCLASSICO PHUNTS PNEWMAN PPREGO PPRIVLABEL PRAGU; run; proc means data=spaghettisauce n mean median std min max; var WCLASSICO WHUNTS WNEWMAN WPREGO WPRIVLABEL WRAGU wcheck; run; 81

82 82 * Seemingly Unrelated Regression LA/AIDS Model; proc syslin data=spaghettisauce itsur; classico: model wclassico=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; hunts: model whunts=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; newman: model wnewman=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; ragu: model wragu=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; prego: model wprego=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw;

83 Homogeneity Restrictions srestrict classico.lpclassico+classico.lphunts+ classico.lpnewman+classico.lpprego+classico. lpprivlabel+classico.lpragu=0, hunts.lpclassico+hunts.lphunts+hunts.lpnewman+hunts. lpprego+hunts.lpprivlabel+hunts.lpragu=0, newman.lpclassico+newman.lphunts+newman.lpnewman+ newman.lpprego+newman.lpprivlabel+newman.lpragu=0, ragu.lpclassico+ragu.lphunts+ragu.lpnewman+ragu. lpprego+ragu.lpprivlabel+ragu.lpragu=0, prego.lpclassico+prego.lphunts+prego.lpnewman+prego. lpprego+prego.lpprivlabel+prego.lpragu=0, 83

84 Symmetry Restrictions classico.lphunts-hunts.lpclassico=0, classico.lpnewman-newman.lpclassico=0, classico.lpragu-ragu.lpclassico=0, classico.lpprego-prego.lpclassico=0, hunts.lpnewman-newman.lphunts=0, hunts.lpragu-ragu.lphunts=0, hunts.lpprego-prego.lphunts=0, newman.lpragu-ragu.lpnewman=0, newman.lpprego-prego.lpnewman=0, ragu.lpprego-prego.lpragu=0; run; 84

85 NOTE: Due to adding-up, one equation must be dropped to avoid the singularity of Σ. In this case, the omitted equation is for private label spaghetti sauce. * system test of homogeneity and symmetry restrictions; proc syslin data=spaghettisauce itsur; classico: model wclassico=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; hunts: model whunts=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; newman: model wnewman=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; ragu: model wragu=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; prego: model wprego=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; 85

86 Test of homogeneity restrictions stest classico.lpclassico+classico.lphunts+classico. lpnewman+classico.lpprego+classico.lpprivlabel+ classico.lpragu=0, hunts.lpclassico+hunts.lphunts+hunts.lpnewman+hunts. lpprego+hunts.lpprivlabel+hunts.lpragu=0, newman.lpclassico+newman.lphunts+newman.lpnewman+newman. lpprego+newman.lpprivlabel+newman.lpragu=0, ragu.lpclassico+ragu.lphunts+ragu.lpnewman+ragu.lpprego+ ragu.lpprivlabel+ragu.lpragu=0, prego.lpclassico+prego.lphunts+prego.lpnewman+prego. lpprego+prego.lpprivlabel+prego.lpragu=0, 86

87 Test of symmetry restrictions classico.lphunts-hunts.lpclassico=0, classico.lpnewman-newman.lpclassico=0, classico.lpragu-ragu.lpclassico=0, classico.lpprego-prego.lpclassico=0, hunts.lpnewman-newman.lphunts=0, hunts.lpragu-ragu.lphunts=0, hunts.lpprego-prego.lphunts=0, newman.lpragu-ragu.lpnewman=0, newman.lpprego-prego.lpnewman=0, ragu.lpprego-prego.lpragu=0; run; * system test of homogeneity restrictions only; 87

88 proc syslin data=spaghettisauce itsur; classico: model wclassico=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; hunts: model whunts=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; newman: model wnewman=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; ragu: model wragu=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; prego: model wprego=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; stest 88

89 Test of homogeneity restrictions only classico.lpclassico+classico.lphunts+classico.lpnewman+ classico.lpprego+classico.lpprivlabel+classico.lpragu=0, hunts.lpclassico+hunts.lphunts+hunts.lpnewman+hunts.lpprego+ hunts.lpprivlabel+hunts.lpragu=0, newman.lpclassico+newman.lphunts+newman.lpnewman+newman. lpprego+newman.lpprivlabel+newman.lpragu=0, 89 ragu.lpclassico+ragu.lphunts+ragu.lpnewman+ragu.lpprego+ragu. lpprivlabel+ragu.lpragu=0, prego.lpclassico+prego.lphunts+prego.lpnewman+prego.lpprego+ prego.lpprivlabel+prego.lpragu=0; run; * system test of symmetry restrictions only;

90 proc syslin data=spaghettisauce itsur; classico: model wclassico=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; hunts: model whunts=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; newman: model wnewman=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; ragu: model wragu=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; prego: model wprego=lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp / dw; stest 90

91 Test of symmetry restrictions only classico.lphunts-hunts.lpclassico=0, classico.lpnewman-newman.lpclassico=0, classico.lpragu-ragu.lpclassico=0, classico.lpprego-prego.lpclassico=0, hunts.lpnewman-newman.lphunts=0, hunts.lpragu-ragu.lphunts=0, hunts.lpprego-prego.lphunts=0, newman.lpragu-ragu.lpnewman=0, newman.lpprego-prego.lpnewman=0, ragu.lpprego-prego.lpragu=0; run; 91

92 The MEANS Procedure Volume of Spaghetti Sauce in standard units (15 oz container size) Variable N Mean Median Std Dev Minimum Maximum ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ QCLASSICO QHUNTS QNEWMAN QPREGO QPRIVLABEL QRAGU ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ The MEANS Procedure Prices/Standardized container Variable N Mean Median Std Dev Minimum Maximum ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ PCLASSICO PHUNTS PNEWMAN PPREGO PPRIVLABEL PRAGU ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ 92

93 The MEANS Procedure Market Shares Variable N Mean Median Std Dev Minimum Maximum ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ wclassico whunts wnewman wprego wprivlabel wragu wcheck E ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ Market shares need to add to 1. Ragu and Prego are the industry leaders 93

94 The SYSLIN Procedure Ordinary Least Squares Estimation (No Restrictions Imposed) Model CLASSICO Dependent Variable wclassico Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

95 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp <

96 The SYSLIN Procedure Ordinary Least Squares Estimation Model HUNTS Dependent Variable whunts Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

97 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept lpclassico lphunts <.0001 lpnewman lpprego lpprivlabel lpragu lrealexp

98 The SYSLIN Procedure Ordinary Least Squares Estimation Model NEWMAN Dependent Variable wnewman Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error E-6 Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

99 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept <.0001 lpclassico lphunts lpnewman <.0001 lpprego lpprivlabel lpragu lrealexp

100 The SYSLIN Procedure Ordinary Least Squares Estimation Model RAGU Dependent Variable wragu Analysis of Variance Sum of Mean Source DF Squares Square F Value Pr > F Model <.0001 Error Corrected Total Root MSE R-Square Dependent Mean Adj R-Sq Coeff Var

101 Parameter Estimates Parameter Standard Variable DF Estimate Error t Value Pr > t Intercept lpclassico lphunts lpnewman lpprego lpprivlabel lpragu lrealexp <

unadjusted model for baseline cholesterol 22:31 Monday, April 19,

unadjusted model for baseline cholesterol 22:31 Monday, April 19, unadjusted model for baseline cholesterol 22:31 Monday, April 19, 2004 1 Class Level Information Class Levels Values TRETGRP 3 3 4 5 SEX 2 0 1 Number of observations 916 unadjusted model for baseline cholesterol

More information

4. Examples. Results: Example 4.1 Implementation of the Example 3.1 in SAS. In SAS we can use the Proc Model procedure.

4. Examples. Results: Example 4.1 Implementation of the Example 3.1 in SAS. In SAS we can use the Proc Model procedure. 4. Examples Example 4.1 Implementation of the Example 3.1 in SAS. In SAS we can use the Proc Model procedure. Simulate data from t-distribution with ν = 6. SAS: data tdist; do i = 1 to 500; y = tinv(ranuni(158),6);

More information

SAS/ETS 14.3 User s Guide The SYSLIN Procedure

SAS/ETS 14.3 User s Guide The SYSLIN Procedure SAS/ETS 14.3 User s Guide The SYSLIN Procedure This document is an individual chapter from SAS/ETS 14.3 User s Guide. The correct bibliographic citation for this manual is as follows: SAS Institute Inc.

More information

CHAPTER 4: Forecasting by Regression

CHAPTER 4: Forecasting by Regression CHAPTER 4: Forecasting by Regression Prof. Alan Wan 1 / 57 Table of contents 1. Revision of Linear Regression 3.1 First-order Autocorrelation and the Durbin-Watson Test 3.2 Correction for Autocorrelation

More information

This is a Randomized Block Design (RBD) with a single factor treatment arrangement (2 levels) which are fixed.

This is a Randomized Block Design (RBD) with a single factor treatment arrangement (2 levels) which are fixed. EXST3201 Chapter 13c Geaghan Fall 2005: Page 1 Linear Models Y ij = µ + βi + τ j + βτij + εijk This is a Randomized Block Design (RBD) with a single factor treatment arrangement (2 levels) which are fixed.

More information

Econometrics. 9) Heteroscedasticity and autocorrelation

Econometrics. 9) Heteroscedasticity and autocorrelation 30C00200 Econometrics 9) Heteroscedasticity and autocorrelation Timo Kuosmanen Professor, Ph.D. http://nomepre.net/index.php/timokuosmanen Today s topics Heteroscedasticity Possible causes Testing for

More information

SAS Procedures Inference about the Line ffl model statement in proc reg has many options ffl To construct confidence intervals use alpha=, clm, cli, c

SAS Procedures Inference about the Line ffl model statement in proc reg has many options ffl To construct confidence intervals use alpha=, clm, cli, c Inference About the Slope ffl As with all estimates, ^fi1 subject to sampling var ffl Because Y jx _ Normal, the estimate ^fi1 _ Normal A linear combination of indep Normals is Normal Simple Linear Regression

More information

Instrumental Variables, Simultaneous and Systems of Equations

Instrumental Variables, Simultaneous and Systems of Equations Chapter 6 Instrumental Variables, Simultaneous and Systems of Equations 61 Instrumental variables In the linear regression model y i = x iβ + ε i (61) we have been assuming that bf x i and ε i are uncorrelated

More information

Detecting and Assessing Data Outliers and Leverage Points

Detecting and Assessing Data Outliers and Leverage Points Chapter 9 Detecting and Assessing Data Outliers and Leverage Points Section 9.1 Background Background Because OLS estimators arise due to the minimization of the sum of squared errors, large residuals

More information

Chapter 2. Appendix A: Vertical line model parameter estimates

Chapter 2. Appendix A: Vertical line model parameter estimates Chapter 2 Appendix A: Vertical line model parameter estimates 1 Model Summary Model Variables 4 Parameters 166 Equations 4 Number of Statements 252 Note: The parameter beta12 is shared by 2 of the equations

More information

Exercices for Applied Econometrics A

Exercices for Applied Econometrics A QEM F. Gardes-C. Starzec-M.A. Diaye Exercices for Applied Econometrics A I. Exercice: The panel of households expenditures in Poland, for years 1997 to 2000, gives the following statistics for the whole

More information

Autocorrelation or Serial Correlation

Autocorrelation or Serial Correlation Chapter 6 Autocorrelation or Serial Correlation Section 6.1 Introduction 2 Evaluating Econometric Work How does an analyst know when the econometric work is completed? 3 4 Evaluating Econometric Work Econometric

More information

5.3 Three-Stage Nested Design Example

5.3 Three-Stage Nested Design Example 5.3 Three-Stage Nested Design Example A researcher designs an experiment to study the of a metal alloy. A three-stage nested design was conducted that included Two alloy chemistry compositions. Three ovens

More information

Lecture 11: Simple Linear Regression

Lecture 11: Simple Linear Regression Lecture 11: Simple Linear Regression Readings: Sections 3.1-3.3, 11.1-11.3 Apr 17, 2009 In linear regression, we examine the association between two quantitative variables. Number of beers that you drink

More information

Romanian Economic and Business Review Vol. 3, No. 3 THE EVOLUTION OF SNP PETROM STOCK LIST - STUDY THROUGH AUTOREGRESSIVE MODELS

Romanian Economic and Business Review Vol. 3, No. 3 THE EVOLUTION OF SNP PETROM STOCK LIST - STUDY THROUGH AUTOREGRESSIVE MODELS THE EVOLUTION OF SNP PETROM STOCK LIST - STUDY THROUGH AUTOREGRESSIVE MODELS Marian Zaharia, Ioana Zaheu, and Elena Roxana Stan Abstract Stock exchange market is one of the most dynamic and unpredictable

More information

Paper: ST-161. Techniques for Evidence-Based Decision Making Using SAS Ian Stockwell, The Hilltop UMBC, Baltimore, MD

Paper: ST-161. Techniques for Evidence-Based Decision Making Using SAS Ian Stockwell, The Hilltop UMBC, Baltimore, MD Paper: ST-161 Techniques for Evidence-Based Decision Making Using SAS Ian Stockwell, The Hilltop Institute @ UMBC, Baltimore, MD ABSTRACT SAS has many tools that can be used for data analysis. From Freqs

More information

General Linear Model (Chapter 4)

General Linear Model (Chapter 4) General Linear Model (Chapter 4) Outcome variable is considered continuous Simple linear regression Scatterplots OLS is BLUE under basic assumptions MSE estimates residual variance testing regression coefficients

More information

Failure Time of System due to the Hot Electron Effect

Failure Time of System due to the Hot Electron Effect of System due to the Hot Electron Effect 1 * exresist; 2 option ls=120 ps=75 nocenter nodate; 3 title of System due to the Hot Electron Effect ; 4 * TIME = failure time (hours) of a system due to drift

More information

Use of Dummy (Indicator) Variables in Applied Econometrics

Use of Dummy (Indicator) Variables in Applied Econometrics Chapter 5 Use of Dummy (Indicator) Variables in Applied Econometrics Section 5.1 Introduction Use of Dummy (Indicator) Variables Model specifications in applied econometrics often necessitate the use of

More information

14.32 Final : Spring 2001

14.32 Final : Spring 2001 14.32 Final : Spring 2001 Please read the entire exam before you begin. You have 3 hours. No books or notes should be used. Calculators are allowed. There are 105 points. Good luck! A. True/False/Sometimes

More information

T-test: means of Spock's judge versus all other judges 1 12:10 Wednesday, January 5, judge1 N Mean Std Dev Std Err Minimum Maximum

T-test: means of Spock's judge versus all other judges 1 12:10 Wednesday, January 5, judge1 N Mean Std Dev Std Err Minimum Maximum T-test: means of Spock's judge versus all other judges 1 The TTEST Procedure Variable: pcwomen judge1 N Mean Std Dev Std Err Minimum Maximum OTHER 37 29.4919 7.4308 1.2216 16.5000 48.9000 SPOCKS 9 14.6222

More information

HOW TO TEST ENDOGENEITY OR EXOGENEITY: AN E-LEARNING HANDS ON SAS

HOW TO TEST ENDOGENEITY OR EXOGENEITY: AN E-LEARNING HANDS ON SAS How to Test Endogeneity or Exogeneity: An E-Learning Hands on SAS 1 HOW TO TEST ENDOGENEITY OR EXOGENEITY: AN E-LEARNING HANDS ON SAS *N. Uttam Singh, **Kishore K Das and *Aniruddha Roy *ICAR Research

More information

Chapter 8 (More on Assumptions for the Simple Linear Regression)

Chapter 8 (More on Assumptions for the Simple Linear Regression) EXST3201 Chapter 8b Geaghan Fall 2005: Page 1 Chapter 8 (More on Assumptions for the Simple Linear Regression) Your textbook considers the following assumptions: Linearity This is not something I usually

More information

Handout 1: Predicting GPA from SAT

Handout 1: Predicting GPA from SAT Handout 1: Predicting GPA from SAT appsrv01.srv.cquest.utoronto.ca> appsrv01.srv.cquest.utoronto.ca> ls Desktop grades.data grades.sas oldstuff sasuser.800 appsrv01.srv.cquest.utoronto.ca> cat grades.data

More information

Financial Econometrics

Financial Econometrics Financial Econometrics Multivariate Time Series Analysis: VAR Gerald P. Dwyer Trinity College, Dublin January 2013 GPD (TCD) VAR 01/13 1 / 25 Structural equations Suppose have simultaneous system for supply

More information

Practical Econometrics. for. Finance and Economics. (Econometrics 2)

Practical Econometrics. for. Finance and Economics. (Econometrics 2) Practical Econometrics for Finance and Economics (Econometrics 2) Seppo Pynnönen and Bernd Pape Department of Mathematics and Statistics, University of Vaasa 1. Introduction 1.1 Econometrics Econometrics

More information

G. S. Maddala Kajal Lahiri. WILEY A John Wiley and Sons, Ltd., Publication

G. S. Maddala Kajal Lahiri. WILEY A John Wiley and Sons, Ltd., Publication G. S. Maddala Kajal Lahiri WILEY A John Wiley and Sons, Ltd., Publication TEMT Foreword Preface to the Fourth Edition xvii xix Part I Introduction and the Linear Regression Model 1 CHAPTER 1 What is Econometrics?

More information

Covariance Structure Approach to Within-Cases

Covariance Structure Approach to Within-Cases Covariance Structure Approach to Within-Cases Remember how the data file grapefruit1.data looks: Store sales1 sales2 sales3 1 62.1 61.3 60.8 2 58.2 57.9 55.1 3 51.6 49.2 46.2 4 53.7 51.5 48.3 5 61.4 58.7

More information

Demand analysis is one of the rst topics come to in economics. Very important especially in the Keynesian paradigm.

Demand analysis is one of the rst topics come to in economics. Very important especially in the Keynesian paradigm. 1 Demand Analysis Demand analysis is one of the rst topics come to in economics. Very important especially in the Keynesian paradigm. Very important for companies: mainstay of consultancies As have seen

More information

Answer all questions from part I. Answer two question from part II.a, and one question from part II.b.

Answer all questions from part I. Answer two question from part II.a, and one question from part II.b. B203: Quantitative Methods Answer all questions from part I. Answer two question from part II.a, and one question from part II.b. Part I: Compulsory Questions. Answer all questions. Each question carries

More information

Auto correlation 2. Note: In general we can have AR(p) errors which implies p lagged terms in the error structure, i.e.,

Auto correlation 2. Note: In general we can have AR(p) errors which implies p lagged terms in the error structure, i.e., 1 Motivation Auto correlation 2 Autocorrelation occurs when what happens today has an impact on what happens tomorrow, and perhaps further into the future This is a phenomena mainly found in time-series

More information

UNIVERSITY OF TORONTO. Faculty of Arts and Science APRIL 2010 EXAMINATIONS STA 303 H1S / STA 1002 HS. Duration - 3 hours. Aids Allowed: Calculator

UNIVERSITY OF TORONTO. Faculty of Arts and Science APRIL 2010 EXAMINATIONS STA 303 H1S / STA 1002 HS. Duration - 3 hours. Aids Allowed: Calculator UNIVERSITY OF TORONTO Faculty of Arts and Science APRIL 2010 EXAMINATIONS STA 303 H1S / STA 1002 HS Duration - 3 hours Aids Allowed: Calculator LAST NAME: FIRST NAME: STUDENT NUMBER: There are 27 pages

More information

Final Exam. 1. Definitions: Briefly Define each of the following terms as they relate to the material covered in class.

Final Exam. 1. Definitions: Briefly Define each of the following terms as they relate to the material covered in class. Name Answer Key Economics 170 Spring 2003 Honor pledge: I have neither given nor received aid on this exam including the preparation of my one page formula list and the preparation of the Stata assignment

More information

Topic 17 - Single Factor Analysis of Variance. Outline. One-way ANOVA. The Data / Notation. One way ANOVA Cell means model Factor effects model

Topic 17 - Single Factor Analysis of Variance. Outline. One-way ANOVA. The Data / Notation. One way ANOVA Cell means model Factor effects model Topic 17 - Single Factor Analysis of Variance - Fall 2013 One way ANOVA Cell means model Factor effects model Outline Topic 17 2 One-way ANOVA Response variable Y is continuous Explanatory variable is

More information

TIME SERIES DATA ANALYSIS USING EVIEWS

TIME SERIES DATA ANALYSIS USING EVIEWS TIME SERIES DATA ANALYSIS USING EVIEWS I Gusti Ngurah Agung Graduate School Of Management Faculty Of Economics University Of Indonesia Ph.D. in Biostatistics and MSc. in Mathematical Statistics from University

More information

Measuring relationships among multiple responses

Measuring relationships among multiple responses Measuring relationships among multiple responses Linear association (correlation, relatedness, shared information) between pair-wise responses is an important property used in almost all multivariate analyses.

More information

Lecture notes on Regression & SAS example demonstration

Lecture notes on Regression & SAS example demonstration Regression & Correlation (p. 215) When two variables are measured on a single experimental unit, the resulting data are called bivariate data. You can describe each variable individually, and you can also

More information

APPLIED MACROECONOMETRICS Licenciatura Universidade Nova de Lisboa Faculdade de Economia. FINAL EXAM JUNE 3, 2004 Starts at 14:00 Ends at 16:30

APPLIED MACROECONOMETRICS Licenciatura Universidade Nova de Lisboa Faculdade de Economia. FINAL EXAM JUNE 3, 2004 Starts at 14:00 Ends at 16:30 APPLIED MACROECONOMETRICS Licenciatura Universidade Nova de Lisboa Faculdade de Economia FINAL EXAM JUNE 3, 2004 Starts at 14:00 Ends at 16:30 I In Figure I.1 you can find a quarterly inflation rate series

More information

Topic 20: Single Factor Analysis of Variance

Topic 20: Single Factor Analysis of Variance Topic 20: Single Factor Analysis of Variance Outline Single factor Analysis of Variance One set of treatments Cell means model Factor effects model Link to linear regression using indicator explanatory

More information

Circle the single best answer for each multiple choice question. Your choice should be made clearly.

Circle the single best answer for each multiple choice question. Your choice should be made clearly. TEST #1 STA 4853 March 6, 2017 Name: Please read the following directions. DO NOT TURN THE PAGE UNTIL INSTRUCTED TO DO SO Directions This exam is closed book and closed notes. There are 32 multiple choice

More information

Analysis of Longitudinal Data: Comparison Between PROC GLM and PROC MIXED. Maribeth Johnson Medical College of Georgia Augusta, GA

Analysis of Longitudinal Data: Comparison Between PROC GLM and PROC MIXED. Maribeth Johnson Medical College of Georgia Augusta, GA Analysis of Longitudinal Data: Comparison Between PROC GLM and PROC MIXED Maribeth Johnson Medical College of Georgia Augusta, GA Overview Introduction to longitudinal data Describe the data for examples

More information

Vector Autoregressive Model. Vector Autoregressions II. Estimation of Vector Autoregressions II. Estimation of Vector Autoregressions I.

Vector Autoregressive Model. Vector Autoregressions II. Estimation of Vector Autoregressions II. Estimation of Vector Autoregressions I. Vector Autoregressive Model Vector Autoregressions II Empirical Macroeconomics - Lect 2 Dr. Ana Beatriz Galvao Queen Mary University of London January 2012 A VAR(p) model of the m 1 vector of time series

More information

AUTOCORRELATION. Phung Thanh Binh

AUTOCORRELATION. Phung Thanh Binh AUTOCORRELATION Phung Thanh Binh OUTLINE Time series Gauss-Markov conditions The nature of autocorrelation Causes of autocorrelation Consequences of autocorrelation Detecting autocorrelation Remedial measures

More information

Chapter 15 Panel Data Models. Pooling Time-Series and Cross-Section Data

Chapter 15 Panel Data Models. Pooling Time-Series and Cross-Section Data Chapter 5 Panel Data Models Pooling Time-Series and Cross-Section Data Sets of Regression Equations The topic can be introduced wh an example. A data set has 0 years of time series data (from 935 to 954)

More information

at least 50 and preferably 100 observations should be available to build a proper model

at least 50 and preferably 100 observations should be available to build a proper model III Box-Jenkins Methods 1. Pros and Cons of ARIMA Forecasting a) need for data at least 50 and preferably 100 observations should be available to build a proper model used most frequently for hourly or

More information

Odor attraction CRD Page 1

Odor attraction CRD Page 1 Odor attraction CRD Page 1 dm'log;clear;output;clear'; options ps=512 ls=99 nocenter nodate nonumber nolabel FORMCHAR=" ---- + ---+= -/\*"; ODS LISTING; *** Table 23.2 ********************************************;

More information

Possibly useful formulas for this exam: b1 = Corr(X,Y) SDY / SDX. confidence interval: Estimate ± (Critical Value) (Standard Error of Estimate)

Possibly useful formulas for this exam: b1 = Corr(X,Y) SDY / SDX. confidence interval: Estimate ± (Critical Value) (Standard Error of Estimate) Statistics 5100 Exam 2 (Practice) Directions: Be sure to answer every question, and do not spend too much time on any part of any question. Be concise with all your responses. Partial SAS output and statistical

More information

dm'log;clear;output;clear'; options ps=512 ls=99 nocenter nodate nonumber nolabel FORMCHAR=" = -/\<>*"; ODS LISTING;

dm'log;clear;output;clear'; options ps=512 ls=99 nocenter nodate nonumber nolabel FORMCHAR= = -/\<>*; ODS LISTING; dm'log;clear;output;clear'; options ps=512 ls=99 nocenter nodate nonumber nolabel FORMCHAR=" ---- + ---+= -/\*"; ODS LISTING; *** Table 23.2 ********************************************; *** Moore, David

More information

Effect of Centering and Standardization in Moderation Analysis

Effect of Centering and Standardization in Moderation Analysis Effect of Centering and Standardization in Moderation Analysis Raw Data The CORR Procedure 3 Variables: govact negemot Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum Label govact 4.58699

More information

Introduction to Eco n o m et rics

Introduction to Eco n o m et rics 2008 AGI-Information Management Consultants May be used for personal purporses only or by libraries associated to dandelon.com network. Introduction to Eco n o m et rics Third Edition G.S. Maddala Formerly

More information

ECON 4551 Econometrics II Memorial University of Newfoundland. Panel Data Models. Adapted from Vera Tabakova s notes

ECON 4551 Econometrics II Memorial University of Newfoundland. Panel Data Models. Adapted from Vera Tabakova s notes ECON 4551 Econometrics II Memorial University of Newfoundland Panel Data Models Adapted from Vera Tabakova s notes 15.1 Grunfeld s Investment Data 15.2 Sets of Regression Equations 15.3 Seemingly Unrelated

More information

Multicollinearity Exercise

Multicollinearity Exercise Multicollinearity Exercise Use the attached SAS output to answer the questions. [OPTIONAL: Copy the SAS program below into the SAS editor window and run it.] You do not need to submit any output, so there

More information

Week 7.1--IES 612-STA STA doc

Week 7.1--IES 612-STA STA doc Week 7.1--IES 612-STA 4-573-STA 4-576.doc IES 612/STA 4-576 Winter 2009 ANOVA MODELS model adequacy aka RESIDUAL ANALYSIS Numeric data samples from t populations obtained Assume Y ij ~ independent N(μ

More information

Econometrics Homework 1

Econometrics Homework 1 Econometrics Homework Due Date: March, 24. by This problem set includes questions for Lecture -4 covered before midterm exam. Question Let z be a random column vector of size 3 : z = @ (a) Write out z

More information

Eco and Bus Forecasting Fall 2016 EXERCISE 2

Eco and Bus Forecasting Fall 2016 EXERCISE 2 ECO 5375-701 Prof. Tom Fomby Eco and Bus Forecasting Fall 016 EXERCISE Purpose: To learn how to use the DTDS model to test for the presence or absence of seasonality in time series data and to estimate

More information

LECTURE 11. Introduction to Econometrics. Autocorrelation

LECTURE 11. Introduction to Econometrics. Autocorrelation LECTURE 11 Introduction to Econometrics Autocorrelation November 29, 2016 1 / 24 ON PREVIOUS LECTURES We discussed the specification of a regression equation Specification consists of choosing: 1. correct

More information

data proc sort proc corr run proc reg run proc glm run proc glm run proc glm run proc reg CONMAIN CONINT run proc reg DUMMAIN DUMINT run proc reg

data proc sort proc corr run proc reg run proc glm run proc glm run proc glm run proc reg CONMAIN CONINT run proc reg DUMMAIN DUMINT run proc reg data one; input id Y group X; I1=0;I2=0;I3=0;if group=1 then I1=1;if group=2 then I2=1;if group=3 then I3=1; IINT1=I1*X;IINT2=I2*X;IINT3=I3*X; *************************************************************************;

More information

Answers: Problem Set 9. Dynamic Models

Answers: Problem Set 9. Dynamic Models Answers: Problem Set 9. Dynamic Models 1. Given annual data for the period 1970-1999, you undertake an OLS regression of log Y on a time trend, defined as taking the value 1 in 1970, 2 in 1972 etc. The

More information

1) Answer the following questions as true (T) or false (F) by circling the appropriate letter.

1) Answer the following questions as true (T) or false (F) by circling the appropriate letter. 1) Answer the following questions as true (T) or false (F) by circling the appropriate letter. T F T F T F a) Variance estimates should always be positive, but covariance estimates can be either positive

More information

9) Time series econometrics

9) Time series econometrics 30C00200 Econometrics 9) Time series econometrics Timo Kuosmanen Professor Management Science http://nomepre.net/index.php/timokuosmanen 1 Macroeconomic data: GDP Inflation rate Examples of time series

More information

Sample Exam Questions for Econometrics

Sample Exam Questions for Econometrics Sample Exam Questions for Econometrics 1 a) What is meant by marginalisation and conditioning in the process of model reduction within the dynamic modelling tradition? (30%) b) Having derived a model for

More information

Outline. Nature of the Problem. Nature of the Problem. Basic Econometrics in Transportation. Autocorrelation

Outline. Nature of the Problem. Nature of the Problem. Basic Econometrics in Transportation. Autocorrelation 1/30 Outline Basic Econometrics in Transportation Autocorrelation Amir Samimi What is the nature of autocorrelation? What are the theoretical and practical consequences of autocorrelation? Since the assumption

More information

Reading Assignment. Serial Correlation and Heteroskedasticity. Chapters 12 and 11. Kennedy: Chapter 8. AREC-ECON 535 Lec F1 1

Reading Assignment. Serial Correlation and Heteroskedasticity. Chapters 12 and 11. Kennedy: Chapter 8. AREC-ECON 535 Lec F1 1 Reading Assignment Serial Correlation and Heteroskedasticity Chapters 1 and 11. Kennedy: Chapter 8. AREC-ECON 535 Lec F1 1 Serial Correlation or Autocorrelation y t = β 0 + β 1 x 1t + β x t +... + β k

More information

Chapter 8 Quantitative and Qualitative Predictors

Chapter 8 Quantitative and Qualitative Predictors STAT 525 FALL 2017 Chapter 8 Quantitative and Qualitative Predictors Professor Dabao Zhang Polynomial Regression Multiple regression using X 2 i, X3 i, etc as additional predictors Generates quadratic,

More information

Econ 300/QAC 201: Quantitative Methods in Economics/Applied Data Analysis. 17th Class 7/1/10

Econ 300/QAC 201: Quantitative Methods in Economics/Applied Data Analysis. 17th Class 7/1/10 Econ 300/QAC 201: Quantitative Methods in Economics/Applied Data Analysis 17th Class 7/1/10 The only function of economic forecasting is to make astrology look respectable. --John Kenneth Galbraith show

More information

A Guide to Modern Econometric:

A Guide to Modern Econometric: A Guide to Modern Econometric: 4th edition Marno Verbeek Rotterdam School of Management, Erasmus University, Rotterdam B 379887 )WILEY A John Wiley & Sons, Ltd., Publication Contents Preface xiii 1 Introduction

More information

Lecture 3 Linear random intercept models

Lecture 3 Linear random intercept models Lecture 3 Linear random intercept models Example: Weight of Guinea Pigs Body weights of 48 pigs in 9 successive weeks of follow-up (Table 3.1 DLZ) The response is measures at n different times, or under

More information

ECON 4230 Intermediate Econometric Theory Exam

ECON 4230 Intermediate Econometric Theory Exam ECON 4230 Intermediate Econometric Theory Exam Multiple Choice (20 pts). Circle the best answer. 1. The Classical assumption of mean zero errors is satisfied if the regression model a) is linear in the

More information

GLS and related issues

GLS and related issues GLS and related issues Bernt Arne Ødegaard 27 April 208 Contents Problems in multivariate regressions 2. Problems with assumed i.i.d. errors...................................... 2 2 NON-iid errors 2 2.

More information

Volatility. Gerald P. Dwyer. February Clemson University

Volatility. Gerald P. Dwyer. February Clemson University Volatility Gerald P. Dwyer Clemson University February 2016 Outline 1 Volatility Characteristics of Time Series Heteroskedasticity Simpler Estimation Strategies Exponentially Weighted Moving Average Use

More information

Outline. Linear OLS Models vs: Linear Marginal Models Linear Conditional Models. Random Intercepts Random Intercepts & Slopes

Outline. Linear OLS Models vs: Linear Marginal Models Linear Conditional Models. Random Intercepts Random Intercepts & Slopes Lecture 2.1 Basic Linear LDA 1 Outline Linear OLS Models vs: Linear Marginal Models Linear Conditional Models Random Intercepts Random Intercepts & Slopes Cond l & Marginal Connections Empirical Bayes

More information

Monday 7 th Febraury 2005

Monday 7 th Febraury 2005 Monday 7 th Febraury 2 Analysis of Pigs data Data: Body weights of 48 pigs at 9 successive follow-up visits. This is an equally spaced data. It is always a good habit to reshape the data, so we can easily

More information

Lecture 8. Using the CLR Model. Relation between patent applications and R&D spending. Variables

Lecture 8. Using the CLR Model. Relation between patent applications and R&D spending. Variables Lecture 8. Using the CLR Model Relation between patent applications and R&D spending Variables PATENTS = No. of patents (in 000) filed RDEP = Expenditure on research&development (in billions of 99 $) The

More information

Repeated Measures Part 2: Cartoon data

Repeated Measures Part 2: Cartoon data Repeated Measures Part 2: Cartoon data /*********************** cartoonglm.sas ******************/ options linesize=79 noovp formdlim='_'; title 'Cartoon Data: STA442/1008 F 2005'; proc format; /* value

More information

ssh tap sas913, sas

ssh tap sas913, sas B. Kedem, STAT 430 SAS Examples SAS8 ===================== ssh xyz@glue.umd.edu, tap sas913, sas https://www.statlab.umd.edu/sasdoc/sashtml/onldoc.htm Multiple Regression ====================== 0. Show

More information

Case of single exogenous (iv) variable (with single or multiple mediators) iv à med à dv. = β 0. iv i. med i + α 1

Case of single exogenous (iv) variable (with single or multiple mediators) iv à med à dv. = β 0. iv i. med i + α 1 Mediation Analysis: OLS vs. SUR vs. ISUR vs. 3SLS vs. SEM Note by Hubert Gatignon July 7, 2013, updated November 15, 2013, April 11, 2014, May 21, 2016 and August 10, 2016 In Chap. 11 of Statistical Analysis

More information

Correlation and Simple Linear Regression

Correlation and Simple Linear Regression Correlation and Simple Linear Regression Sasivimol Rattanasiri, Ph.D Section for Clinical Epidemiology and Biostatistics Ramathibodi Hospital, Mahidol University E-mail: sasivimol.rat@mahidol.ac.th 1 Outline

More information

Introduction to Regression Analysis. Dr. Devlina Chatterjee 11 th August, 2017

Introduction to Regression Analysis. Dr. Devlina Chatterjee 11 th August, 2017 Introduction to Regression Analysis Dr. Devlina Chatterjee 11 th August, 2017 What is regression analysis? Regression analysis is a statistical technique for studying linear relationships. One dependent

More information

Section 2 NABE ASTEF 65

Section 2 NABE ASTEF 65 Section 2 NABE ASTEF 65 Econometric (Structural) Models 66 67 The Multiple Regression Model 68 69 Assumptions 70 Components of Model Endogenous variables -- Dependent variables, values of which are determined

More information

Econometrics I KS. Module 2: Multivariate Linear Regression. Alexander Ahammer. This version: April 16, 2018

Econometrics I KS. Module 2: Multivariate Linear Regression. Alexander Ahammer. This version: April 16, 2018 Econometrics I KS Module 2: Multivariate Linear Regression Alexander Ahammer Department of Economics Johannes Kepler University of Linz This version: April 16, 2018 Alexander Ahammer (JKU) Module 2: Multivariate

More information

Frequency Forecasting using Time Series ARIMA model

Frequency Forecasting using Time Series ARIMA model Frequency Forecasting using Time Series ARIMA model Manish Kumar Tikariha DGM(O) NSPCL Bhilai Abstract In view of stringent regulatory stance and recent tariff guidelines, Deviation Settlement mechanism

More information

Lecture 8. Using the CLR Model

Lecture 8. Using the CLR Model Lecture 8. Using the CLR Model Example of regression analysis. Relation between patent applications and R&D spending Variables PATENTS = No. of patents (in 1000) filed RDEXP = Expenditure on research&development

More information

Applied Econometrics. Applied Econometrics. Applied Econometrics. Applied Econometrics. What is Autocorrelation. Applied Econometrics

Applied Econometrics. Applied Econometrics. Applied Econometrics. Applied Econometrics. What is Autocorrelation. Applied Econometrics Autocorrelation 1. What is 2. What causes 3. First and higher orders 4. Consequences of 5. Detecting 6. Resolving Learning Objectives 1. Understand meaning of in the CLRM 2. What causes 3. Distinguish

More information

Multivariate Analysis of Variance

Multivariate Analysis of Variance Multivariate Analysis of Variance 1 Multivariate Analysis of Variance Objective of Multivariate Analysis of variance (MANOVA) is to determine whether the differences on criterion or dependent variables

More information

Econometrics Lecture 10: Applied Demand Analysis

Econometrics Lecture 10: Applied Demand Analysis Econometrics Lecture 10: Applied Demand Analysis R. G. Pierse 1 Introduction In this lecture we look at the estimation of systems of demand equations. Demand equations were some of the earliest economic

More information

SplineLinear.doc 1 # 9 Last save: Saturday, 9. December 2006

SplineLinear.doc 1 # 9 Last save: Saturday, 9. December 2006 SplineLinear.doc 1 # 9 Problem:... 2 Objective... 2 Reformulate... 2 Wording... 2 Simulating an example... 3 SPSS 13... 4 Substituting the indicator function... 4 SPSS-Syntax... 4 Remark... 4 Result...

More information

Stat 302 Statistical Software and Its Applications SAS: Simple Linear Regression

Stat 302 Statistical Software and Its Applications SAS: Simple Linear Regression 1 Stat 302 Statistical Software and Its Applications SAS: Simple Linear Regression Fritz Scholz Department of Statistics, University of Washington Winter Quarter 2015 February 16, 2015 2 The Spirit of

More information

Dynamic Time Series Regression: A Panacea for Spurious Correlations

Dynamic Time Series Regression: A Panacea for Spurious Correlations International Journal of Scientific and Research Publications, Volume 6, Issue 10, October 2016 337 Dynamic Time Series Regression: A Panacea for Spurious Correlations Emmanuel Alphonsus Akpan *, Imoh

More information

Chapter 9. Multivariate and Within-cases Analysis. 9.1 Multivariate Analysis of Variance

Chapter 9. Multivariate and Within-cases Analysis. 9.1 Multivariate Analysis of Variance Chapter 9 Multivariate and Within-cases Analysis 9.1 Multivariate Analysis of Variance Multivariate means more than one response variable at once. Why do it? Primarily because if you do parallel analyses

More information

STATISTICS 174: APPLIED STATISTICS FINAL EXAM DECEMBER 10, 2002

STATISTICS 174: APPLIED STATISTICS FINAL EXAM DECEMBER 10, 2002 Time allowed: 3 HOURS. STATISTICS 174: APPLIED STATISTICS FINAL EXAM DECEMBER 10, 2002 This is an open book exam: all course notes and the text are allowed, and you are expected to use your own calculator.

More information

Econometrics of Panel Data

Econometrics of Panel Data Econometrics of Panel Data Jakub Mućk Meeting # 2 Jakub Mućk Econometrics of Panel Data Meeting # 2 1 / 26 Outline 1 Fixed effects model The Least Squares Dummy Variable Estimator The Fixed Effect (Within

More information

Applied Multivariate Analysis

Applied Multivariate Analysis Department of Mathematics and Statistics, University of Vaasa, Finland Spring 2017 Discriminant Analysis Background 1 Discriminant analysis Background General Setup for the Discriminant Analysis Descriptive

More information

Testing Symmetry and Homogeneity in the. AIDS with Cointegrated Data Using. Fully-modified Estimation and the Bootstrap.

Testing Symmetry and Homogeneity in the. AIDS with Cointegrated Data Using. Fully-modified Estimation and the Bootstrap. Testing Symmetry and Homogeneity in the AIDS with Cointegrated Data Using Fully-modified Estimation and the Bootstrap. Richard Tiffin and Kelvin Balcombe University of Reading and Imperial College Paper

More information

SEEMINGLY UNRELATED REGRESSION

SEEMINGLY UNRELATED REGRESSION TERM PAPER On SEEMINGLY UNRELATED REGRESSION By Anomita Ghosh Nitin Kumar Sinha Sudipta Ghosh Udayan Rathore ECONOMETRICS II Instructor Mr. Subrata Sarkar INTRODUCTION The SUR system proposed by Arnold

More information

388 Index Differencing test ,232 Distributed lags , 147 arithmetic lag.

388 Index Differencing test ,232 Distributed lags , 147 arithmetic lag. INDEX Aggregation... 104 Almon lag... 135-140,149 AR(1) process... 114-130,240,246,324-325,366,370,374 ARCH... 376-379 ARlMA... 365 Asymptotically unbiased... 13,50 Autocorrelation... 113-130, 142-150,324-325,365-369

More information

NATCOR Regression Modelling for Time Series

NATCOR Regression Modelling for Time Series Universität Hamburg Institut für Wirtschaftsinformatik Prof. Dr. D.B. Preßmar Professor Robert Fildes NATCOR Regression Modelling for Time Series The material presented has been developed with the substantial

More information

Outline. Review regression diagnostics Remedial measures Weighted regression Ridge regression Robust regression Bootstrapping

Outline. Review regression diagnostics Remedial measures Weighted regression Ridge regression Robust regression Bootstrapping Topic 19: Remedies Outline Review regression diagnostics Remedial measures Weighted regression Ridge regression Robust regression Bootstrapping Regression Diagnostics Summary Check normality of the residuals

More information

****Lab 4, Feb 4: EDA and OLS and WLS

****Lab 4, Feb 4: EDA and OLS and WLS ****Lab 4, Feb 4: EDA and OLS and WLS ------- log: C:\Documents and Settings\Default\Desktop\LDA\Data\cows_Lab4.log log type: text opened on: 4 Feb 2004, 09:26:19. use use "Z:\LDA\DataLDA\cowsP.dta", clear.

More information

Nonlinear Regression. Summary. Sample StatFolio: nonlinear reg.sgp

Nonlinear Regression. Summary. Sample StatFolio: nonlinear reg.sgp Nonlinear Regression Summary... 1 Analysis Summary... 4 Plot of Fitted Model... 6 Response Surface Plots... 7 Analysis Options... 10 Reports... 11 Correlation Matrix... 12 Observed versus Predicted...

More information

ANOVA Longitudinal Models for the Practice Effects Data: via GLM

ANOVA Longitudinal Models for the Practice Effects Data: via GLM Psyc 943 Lecture 25 page 1 ANOVA Longitudinal Models for the Practice Effects Data: via GLM Model 1. Saturated Means Model for Session, E-only Variances Model (BP) Variances Model: NO correlation, EQUAL

More information