各位大大们，我有关于STATA面板数据的问题想请教一下，我是一个STATA面板数据小白，最近正在写毕业论文，是关于人均卫生费用的，这是一组2014—2016年的各省人均卫生费用和社会人口特征指标的面板数据，我对两边取了对数，这样系数就是弹性值大小，我在做固定效应的时候，进行F检验，明显结果是显著的，因此fe优于混合回归，但是在通过LSDV法考虑到聚类稳健标准误的时候，个体虚拟变量都不显著，不存在个体效应，这与前面的结论就相矛盾了。那么到底用混合回归还是fe呢？麻烦各位大大们指教，感激！
xtset diqu year
panel variable: diqu (strongly balanced)
time variable: year, 1 to 3
delta: 1 unit
.
end of do-file
. xtreg ln_Y ln_GDP RUI CR OR AR UR BR DR ln_Xp ln_Xb Xth ln_Xad ln_Xan,fe r
Fixed-effects (within) regression Number of obs = 93
Group variable: diqu Number of groups = 31
R-sq: Obs per group:
within = 0.9132 min = 3
between = 0.6662 avg = 3.0
overall = 0.6842 max = 3
F(13,30) = 47.70
corr(u_i, Xb) = -0.5380 Prob > F = 0.0000
(Std. Err. adjusted for 31 clusters in diqu)
------------------------------------------------------------------------------
| Robust
ln_Y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ln_GDP | .3427626 .1569972 2.18 0.037 .0221316 .6633937
RUI | -.15568 .1463131 -1.06 0.296 -.4544912 .1431312
CR | .0175157 .010564 1.66 0.108 -.0040588 .0390903
OR | .006748 .0068859 0.98 0.335 -.007315 .0208109
AR | -.0044527 .0010305 -4.32 0.000 -.0065573 -.0023481
UR | .014255 .0121155 1.18 0.249 -.0104881 .0389981
BR | -.0063933 .0106048 -0.60 0.551 -.0280512 .0152646
DR | -.0416078 .0334749 -1.24 0.224 -.1099726 .0267571
ln_Xp | .6317926 .5127544 1.23 0.227 -.4153917 1.678977
ln_Xb | .7317934 .4116985 1.78 0.086 -.1090071 1.572594
Xth | .5748219 1.012033 0.57 0.574 -1.492026 2.64167
ln_Xad | -.4352201 .3126507 -1.39 0.174 -1.073738 .2032977
ln_Xan | .0018913 .0223364 0.08 0.933 -.0437257 .0475083
_cons | 2.433706 1.917951 1.27 0.214 -1.483273 6.350684
-------------+----------------------------------------------------------------
sigma_u | .21259539
sigma_e | .04357405
rho | .9596841 (fraction of variance due to u_i)
------------------------------------------------------------------------------
. xtreg ln_Y ln_GDP RUI CR OR AR UR BR DR ln_Xp ln_Xb Xth ln_Xad ln_Xan,fe
Fixed-effects (within) regression Number of obs = 93
Group variable: diqu Number of groups = 31
R-sq: Obs per group:
within = 0.9132 min = 3
between = 0.6662 avg = 3.0
overall = 0.6842 max = 3
F(13,49) = 39.67
corr(u_i, Xb) = -0.5380 Prob > F = 0.0000
------------------------------------------------------------------------------
ln_Y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ln_GDP | .3427626 .1523382 2.25 0.029 .0366276 .6488977
RUI | -.15568 .1389986 -1.12 0.268 -.4350083 .1236482
CR | .0175157 .0087636 2.00 0.051 -.0000954 .0351269
OR | .006748 .0080316 0.84 0.405 -.0093921 .022888
AR | -.0044527 .0060658 -0.73 0.466 -.0166425 .0077371
UR | .014255 .0105507 1.35 0.183 -.0069475 .0354574
BR | -.0063933 .0105045 -0.61 0.546 -.0275029 .0147163
DR | -.0416078 .0345705 -1.20 0.235 -.1110799 .0278643
ln_Xp | .6317926 .4115255 1.54 0.131 -.195199 1.458784
ln_Xb | .7317934 .3627123 2.02 0.049 .0028957 1.460691
Xth | .5748219 .9821435 0.59 0.561 -1.398869 2.548513
ln_Xad | -.4352201 .266281 -1.63 0.109 -.9703319 .0998917
ln_Xan | .0018913 .0233671 0.08 0.936 -.0450666 .0488492
_cons | 2.433706 1.687465 1.44 0.156 -.9573819 5.824794
-------------+----------------------------------------------------------------
sigma_u | .21259539
sigma_e | .04357405
rho | .9596841 (fraction of variance due to u_i)
------------------------------------------------------------------------------
F test that all u_i=0: F(30, 49) = 29.77 Prob > F = 0.0000
. reg ln_Y ln_GDP RUI CR OR AR UR BR DR ln_Xp ln_Xb Xth ln_Xad ln_Xan i.diqu,vce(cluster diqu)
Linear regression Number of obs = 93
F(12, 30) = .
Prob > F = .
R-squared = 0.9902
Root MSE = .04357
(Std. Err. adjusted for 31 clusters in diqu)
------------------------------------------------------------------------------
| Robust
ln_Y | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
ln_GDP | .3427626 .1993459 1.72 0.096 -.0643561 .7498814
RUI | -.15568 .1857799 -0.84 0.409 -.5350932 .2237332
CR | .0175157 .0134135 1.31 0.202 -.0098784 .0449098
OR | .006748 .0087434 0.77 0.446 -.0111084 .0246043
AR | -.0044527 .0013085 -3.40 0.002 -.007125 -.0017804
UR | .014255 .0153835 0.93 0.362 -.0171624 .0456723
BR | -.0063933 .0134654 -0.47 0.638 -.0338933 .0211067
DR | -.0416078 .0425045 -0.98 0.335 -.1284134 .0451979
ln_Xp | .6317926 .6510659 0.97 0.340 -.6978613 1.961446
ln_Xb | .7317934 .5227509 1.40 0.172 -.3358064 1.799393
Xth | .5748219 1.285021 0.45 0.658 -2.049542 3.199186
ln_Xad | -.4352201 .3969857 -1.10 0.282 -1.245973 .3755329
ln_Xan | .0018913 .0283615 0.07 0.947 -.0560305 .0598131
|
diqu |
云南 | .2650767 .737557 0.36 0.722 -1.241216 1.771369
内蒙古 | -.1013112 .4749651 -0.21 0.833 -1.071319 .8686969
北京 | .0644995 .2772727 0.23 0.818 -.5017668 .6307658
吉林 | .0071111 .4858147 0.01 0.988 -.985055 .9992771
四川 | .0443033 .6490875 0.07 0.946 -1.28131 1.369917
天津 | -.0021039 .1307192 -0.02 0.987 -.269068 .2648603
宁夏 | .0627205 .5930672 0.11 0.916 -1.148484 1.273925
安徽 | .1409622 .5587505 0.25 0.803 -1.000159 1.282083
山东 | -.2102991 .5147605 -0.41 0.686 -1.26158 .8409822
山西 | .0506364 .5175916 0.10 0.923 -1.006427 1.107699
广东 | -.1596913 .316294 -0.50 0.617 -.8056498 .4862671
广西 | -.1565228 .6536391 -0.24 0.812 -1.491432 1.178386
新疆 | -.0437507 .7976169 -0.05 0.957 -1.672702 1.5852
江苏 | -.1436194 .3572346 -0.40 0.691 -.8731897 .5859509
江西 | -.1129251 .5783743 -0.20 0.847 -1.294123 1.068273
河北 | .0052849 .5743157 0.01 0.993 -1.167624 1.178194
河南 | -.1429388 .6778855 -0.21 0.834 -1.527366 1.241488
浙江 | -.1156606 .3325443 -0.35 0.730 -.7948068 .5634855
海南 | .0575621 .4962598 0.12 0.908 -.9559357 1.07106
湖北 | -.2616069 .5029749 -0.52 0.607 -1.288819 .7656049
湖南 | -.1573022 .6413948 -0.25 0.808 -1.467205 1.152601
甘肃 | .4483938 .7048352 0.64 0.529 -.9910717 1.887859
福建 | -.2115062 .4137992 -0.51 0.613 -1.056597 .6335844
西藏 | .8102299 .9689262 0.84 0.410 -1.168581 2.789041
贵州 | -.0250878 .8306248 -0.03 0.976 -1.72145 1.671274
辽宁 | -.1540108 .3538911 -0.44 0.667 -.8767528 .5687312
重庆 | -.0609141 .4889301 -0.12 0.902 -1.059443 .9376143
陕西 | .0431783 .6014672 0.07 0.943 -1.185181 1.271538
青海 | .2224064 .6905799 0.32 0.750 -1.187946 1.632759
黑龙江 | .0532354 .4199736 0.13 0.900 -.8044652 .9109359
|
_cons | 2.426727 2.40718 1.01 0.321 -2.489391 7.342844
------------------------------------------------------------------------------
扫码加好友，拉您进群
全部版块
我的主页
收藏
