Survey
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project
Limitations of simple regression model: Simultaneous equations 1 Causal model with reciprocal effects U1 P = price D = demand I = Income W = Wages W I D + - U2 P 2 True value a=-.2 Path Diagram V2 *a Y1 Y3 Y2 Y4 Y5 Y6 3 Example SEM, Monte Carlo data /MATRIX 2.5123 0.9345 0.5414 1.4768 0.4544 2.1110 0.6925 1.5067 0.5093 0.3683 0.4155 1.5910 1.4842 2.1037 0.5278 1.4727 1.5566 -0.0683 0.0235 0.0847 0.9376 n = 500 Example SEM, Monte Carlo DAta Data generating process (Matlab code): n = 500; B = [0 .9 .5 0 .5 0; -.2 0 0 .5 0 .5; 0 0 0 0 0 0; 0 0 0 0 0 0; 0 0 0 0 0 0; 0 0 0 0 0 0] IB = inv( eye(6,6)-B); PHI = [.2 0 0 0 0 0; 0 .2 0 0 0 0; 0 0 1 .4 0 0; 0 0 .4 1 .4 0; 0 0 0 .4 1 0; 0 0 0 0 0 1] z = IB*sqrt(PHI)*randn(6, n); SIG = cov(z') 4 Standard OLS regression When ignoring simultaneous equations, i.e. OLS: MEASUREMENT EQUATIONS WITH STANDARD ERRORS AND TEST STATISTICS V2 =V2 = .335*V1 .028 11.899 + -.011*V4 .030 -.365 + .312*V6 .018 17.050 + 1.000 D2 5 SEM analysis EQS analysis: GOODNESS OF FIT SUMMARY /TITLE example of SEM /SPECIFICATIONS CASES = 500 ; VAR = 6; /EQUATIONS V1 = .5*V2 + *V3 + *V5 +D1 ; V2 = -.5*V1 + *V4 + *V6 + D2 ; /VARIANCES D1 = *; D2 = *; V3 = *; V4 = *; V5 = *; V6=*; /COVARIANCES V3 to V6 = *; /MATRIX 2.5123 0.9345 0.5414 1.4768 0.4544 1.5910 2.1110 0.6925 1.4842 2.1037 1.5067 0.5093 0.5278 1.4727 0.3683 0.4155 -0.0683 0.0235 0.9376 /END 1.5566 0.0847 6 Simultaneous Equations MEASUREMENT EQUATIONS WITH STANDARD ERRORS AND TEST STATISTICS V1 =V1 = .863*V2 .046 18.757 + .512*V3 .019 26.765 V2 =V2 = -.146*V1 .058 -2.541 + .471*V4 .059 7.983 + .512*V5 .021 24.445 + 1.000 D1 + + 1.000 D2 .489*V6 .028 17.230 CHI-SQUARE = 0.213 BASED ON 3 DEGREES OF FREEDOM PROBABILITY VALUE FOR THE CHI-SQUARE STATISTIC IS 0.97544 7