用estat hettest检测异方差,p值是0.0772,是不是说明在95%的置信区间不拒绝原假设,就是无异方差?
estat hettest
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: fitted values of lprice
chi2(1) = 3.12
Prob > chi2 = 0.0772
而且我用残差平方和自变量的平方做回归,
reg uhatsq rooms baths larea age agesq y81
Source | SS df MS Number of obs = 321
-------------+---------------------------------- F(6, 314) = 1.52
Model | .12660501 6 .021100835 Prob > F = 0.1726
Residual | 4.37313764 314 .01392719 R-squared = 0.0281
-------------+---------------------------------- Adj R-squared = 0.0096
Total | 4.49974265 320 .014061696 Root MSE = .11801
得到的这个F检测的p值和前边BP检测差很多,
再用LM statistic检测
scalar LM = e(r2)*e(N)
. scalar pvalue = chi2tail(e(df_m),LM)
. disp "Breusch-Pagan test: LM = " LM ", p-value = " pvalue
Breusch-Pagan test: LM = 9.0316739, p-value = .17180457
这个结果也和前面的结果不一样,但是和F检测倒不是差很多。
这是为什么呢?
代码是这样的:
reg lprice rooms baths larea age agesq y81
predict uhat, residual
gen uhatsq=uhat^2
reg uhatsq rooms baths larea age agesq y81
scalar LM = e(r2)*e(N)
scalar pvalue = chi2tail(e(df_m),LM)
disp "Breusch-Pagan test: LM = " LM ", p-value = " pvalue
求解答,谢谢大家。