我用了双重差分分析方法做回归,查了网上的一些方法,但是stata自动给我忽略了一个变量,就是那个交互项,不知道怎么回事,请教请教啊~
tsset vege time
panel variable: vege, 1 to 5
time variable: time, 20080107 to 20120409, but with gaps
. gen i_a=insurance*apply
. xtreg price insurance apply i_a,re
note: i_a dropped due to collinearity(注意这一句!)
Random-effects GLS regression Number of obs = 1085
Group variable (i): vege Number of groups = 5
R-sq: within = 0.0298 Obs per group: min = 193
between = 0.3754 avg = 217.0
overall = 0.0979 max = 223
Random effects u_i ~ Gaussian Wald chi2(2) = 48.26
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000
------------------------------------------------------------------------------
price | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
insurance | 77.09969 14.42446 5.35 0.000 48.82827 105.3711
apply | -149.8199 30.48361 -4.91 0.000 -209.5667 -90.07316
_cons | 410.5202 27.12991 15.13 0.000 357.3465 463.6938
-------------+----------------------------------------------------------------
sigma_u | 24.450014
sigma_e | 164.96337
rho | .02149542 (fraction of variance due to u_i)
------------------------------------------------------------------------------