交通计画ATransportationPlanningA-STRePinIRIDeS.pptVIP

Rにおける尤度関数の定義と最大化 #対数尤度関数の定義　係数ベクトルｘの関数と見なす． LL - function (x){ vbus - x[1] * tb + x[2] * cb vcar - x[1] * tc + x[2] * cc + x[3] ppb - 1/(1+exp(vcar - vbus)) ppc - 1- ppb return(sum(pb*log(ppb)+pc*log(ppc))) } #Optim関数で最大化を行い，結果をresに代入する．初期値は(0,0,0) b0=c(0,0,0) res-optim(b0,LL, method = BFGS, hessian = TRUE, control=list(fnscale=-1)) Rによる推定結果の表示 res $par [1] -0.081037584 -0.004007811 0.369193890 $value [1] -5.047779 $counts function gradient 62 18 $convergence [1] 0 $message NULL $hessian [,1] [,2] [,3] [1,] -19.628306 -613.3939 5.313007 [2,] -613.393875 -23980.3173 196.530577 [3,] 5.313007 196.5306 -1.681972 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * 各個人の選択を説明する場合Rにより，個人ごとのデータを作成 nb - c(39,11,16,22,31,15,21,25,50) nc - c(104,28,51,61,94,63,62,73,155) nn - nb + nc nnf -numeric(length=10) for (i in 1:9){ nnf[i+1]=nnf[i]+nn[i] } chc - numeric(length=nnf[10]-1) tim - numeric(length=nnf[10]-1) cst - numeric(length=nnf[10]-1) for (i in 1:9){ chc[nnf[i]:(nnf[i]+nn[i]-1)] - c(rep(1,nb[i]), rep(0,nc[i])) tim[nnf[i]:(nnf[i]+nn[i]-1)] - rep((tb[i]-tc[i]),nn[i]) cst[nnf[i]:(nnf[i]+nn[i]-1)] - rep((cb[i]-cc[i]),nn[i]) } 各個人の選択を説明する場合pure disaggregate data ans3 - glm(chc ~ tim+cst, family=binomial(link=logit)) summary(ans3) Call: glm(formula = chc ~ tim + cst, family = binomial(link = logit)) Deviance Residuals: Min 1Q Median 3Q Max -0.8215 -0.7630 -0.7470 -0.1019 1.8016 Coefficients: Estimate Std. Error z value Pr(|z|) (Intercept) -0.385889 0.507795 -0.760 0.447 tim -0.079514 0.061803 -1.287 0.198 cst -0.003873 0.003465 -1.118 0.264 (Dispersion parameter for binomial family taken to be 1) Null deviance: 1034.7 on 919 degrees of freedom Residual deviance: 1032.4 on 917 degrees of freedom AIC: 1038.4 Number of Fisher Scoring iterations: 4 理論的補足ランダム項がある場合の選