MATLAB 空间计量模型的实现 [推广有奖]

101楼

kujikawa(未真实交易用户) 发表于 2018-12-29 15:56:20

以空间滞后模型的个体效应为例，程序如下。
具体遇到的问题有：1、LR检验显示错误 2、LM检验有的不显著 3、有些变量的系数与预期相反

T=4;
N=25;
W=normw(W1);
y=A(:,[3]);
x=A(:,[4:20]);

info.lflag=0;
info.model=1;
info.fe=0;
results=sem_panel_FE(y,x,W,T,info);
vnames=strvcat('教育投资','对子女的期望','调整后家庭纯收入','受教育年限','个人收入','子女性别','城乡分类','家庭中学历最高性别','户口类型','国有/集体企业','民办企业','其他类型企业','职业其他','事业单位','私营企业/个体工商','职业为自己/自家干活','外商港澳台投资','ZF部门党政机关');
prt_spnew(results,vnames,1)；

blagre=[results.beta(2:end);results.rho]; % exclude constant
covblagre=results.cov(2:end,2:end);
%LM检验和稳健的LM检验
% Print out effects estimates
LMsarsem_panel(results,W,y,x);
blagfe=[results.beta;results.rho];
covblagfe=results.cov;
% LR-test for joint significance spatial fixed effects
logliklagfe=results.lik;
info.model=1;
results=sar_panel_FE(y,x,W,T,info);
logliklag=results.lik;
LR=-2*(logliklag-logliklagfe);
dof=N;
probability=1-chis_prb(LR,dof);
% Note: probability > 0.05 implies rejection of spatial fixed effects
fprintf(1,'LR-test joint significance spatial fixed effects, degrees of freedom and probability = %9.4f,%6d,%9.4f \n',LR,dof,probability);
% Hausman test FE versus RE of model + spatially lagged dependent variable
covblagfe= covblagfe(1:17,1:17)
blagfe=blagfe(1:17,:)
hausman=(blagfe-blagre)'*inv(covblagre-covblagfe)*(blagfe-blagre);
dof=length(blagfe);
probability=1-chis_prb(abs(hausman),dof);
% Note: probability > 0.05 implies rejection of random effects model in favor of fixed effects model
fprintf(1,'Hausman test-statistic, degrees of freedom and probability = %9.4f,%6d,%9.4f \n',hausman,dof,probability);

%运行结果

Pooled model with spatial error autocorrelation and spatial fixed effects
Dependent Variable =    教育投资
R-squared    = 0.7305
corr-squared = 0.4389
sigma^2       = 0.0096
log-likelihood  =       100.10764
Nobs,Nvar,#FE = 100, 17, 42
# iterations =    14
min and max rho = -0.9900, 0.9900
total time in secs = 1.2800
time for optimiz = 0.5700
time for lndet    = 0.0570
time for t-stats = 0.0660
No lndet approximation used
***************************************************************
Variable       Coefficient  Asymptot t-stat z-probability
对子女的期望          0.228279       0.738370       0.460289
调整后家庭纯收入          1.517292       4.004764       0.000062
受教育年限             0.164816       0.954345       0.339909
个人收入             0.195996       0.879214       0.379285
子女性别             0.128975       0.908650       0.363535
城乡分类             0.079042       0.773469       0.439245
家庭中学历最高性别       -0.084279       -0.751456       0.452378
户口类型             -0.129711       -0.944824       0.344749
国有/集体企业          -0.746339       -2.012652       0.044151
民办企业             -0.630928       -0.844520       0.398379
其他类型企业          -0.651729       -1.576001       0.115026
职业其他             -0.398841       -0.674792       0.499808
事业单位             -0.075951       -0.185438       0.852886
私营企业/个体工商       -0.382527       -1.330488       0.183358
职业为自己/自家干活       -0.511331       -1.568177       0.116840
外商港澳台投资          0.506747       1.008832       0.313055
ZF部门党政机关       -0.164558       -0.318135       0.750382
spat.aut.       0.440971       4.335554       0.000015

LM test no spatial lag, probability       = 23.8529, 0.000
robust LM test no spatial lag, probability = 4.1537, 0.042
LM test no spatial error, probability       = 20.8152, 0.000
robust LM test no spatial error, probability = 1.1159, 0.291
LR-test joint significance spatial fixed effects, degrees of freedom and probability =    NaN, 25,    NaN