. tsset year, yearly time variable: year, 1959 to 1994 delta: 1 year. . reg lhous ldpi lrealp

Transcription

1 - opened on: 24 Mar 2012, 21:29:52. use "G:\stata\( )\demand_2011.dta", clear. de - variable name type format label variable label - year float %ty year time float %9.0g trend pop float %9.0g population, measured in million pi float %9.0g aggregate personal income dpi float %9.0g aggr disposable pers income ptpe float %9.0g price total private expenditure hous float %9.0g housing services phous float %9.0g price housing services realp float %9.0g phous/ptpe*100 lhous float %9.0g log(hous) ldpi float %9.0g log(dpi) lrealp float %9.0g log(realp) - Sorted by: year. tsset year, yearly time variable: year, 1959 to 1994 delta: 1 year. reg lhous ldpi lrealp Source SS df MS Number of obs = F( 2, 33) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi _cons Durbin-Watson d-statistic( 3, 36) = reg lhous L.ldpi L.lrealp F( 2, 32) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE =.01412

2 ldpi L L _cons Durbin-Watson d-statistic( 3, 35) = reg lhous L2.ldpi L2.lrealp Source SS df MS Number of obs = F( 2, 31) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi L L _cons Durbin-Watson d-statistic( 3, 34) = reg lhous ldpi L.ldpi lrealp L.lrealp F( 4, 30) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi L L

3 _cons Durbin-Watson d-statistic( 5, 35) = reg lhous ldpi lrealp L.lhous F( 3, 31) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi lhous L _cons Durbin-Watson d-statistic( 4, 35) = prais lhous ldpi lrealp, corc twostep Iteration 0: rho = Iteration 1: rho = Cochrane-Orcutt AR(1) regression -- twostep estimates F( 2, 32) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi _cons rho

4 Durbin-Watson statistic (transformed) prais lhous ldpi lrealp Iteration 0: rho = Iteration 1: rho = Iteration 2: rho = ( ) Iteration 21: rho = Iteration 22: rho = Prais-Winsten AR(1) regression -- iterated estimates Source SS df MS Number of obs = F( 2, 33) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi _cons rho Durbin-Watson statistic (transformed) prais lhous ldpi lrealp, sse Iteration 1: rho = , criterion = Iteration 2: rho = , criterion = ( ) Iteration 13: rho = , criterion = Iteration 14: rho = , criterion = Prais-Winsten AR(1) regression -- SSE search estimates Source SS df MS Number of obs = F( 2, 33) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi _cons rho Durbin-Watson statistic (transformed) prais lhous ldpi lrealp, corc twostep Iteration 0: rho = Iteration 1: rho = Cochrane-Orcutt AR(1) regression -- twostep estimates

5 F( 2, 32) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi _cons rho Durbin-Watson statistic (transformed) scalar rssr = e(rss). reg lhous ldpi L.ldpi lrealp L.lrealp L.lhous F( 5, 29) = Model Prob > F = Residual R-squared = Adj R-squared = Total Root MSE = ldpi L L lhous L _cons scalar rssu = e(rss). matrix varcov = e(v). scalar varlag = varcov[5,5]. scalar nobs = e(n) scalar list nobs = 35 varlag = rssu = rssr = scalar comfac = nobs*log(rssr/rssu). scalar pvalue = chi2tail(2,comfac)

6 . scalar list pvalue = 7.557e-15 comfac = nobs = 35 varlag = rssu = rssr = log close closed on: 24 MAR 2012, 21:49:41 -