[统计软件] stata做固定效应模型和随机效应模型回归结果一样怎么办 [推广有奖]

1关注
2粉丝

硕士生

55%

还不是VIP/贵宾

威望: 0 级
论坛币: 2306 个
通用积分: 4.8666
学术水平: 0 点
热心指数: 0 点
信用等级: 0 点
经验: 1082 点
帖子: 82
精华: 0
在线时间: 201 小时
注册时间: 2017-3-8
最后登录: 2024-10-11

楼主

hejiyan 发表于 2019-3-5 00:18:32 |只看作者 |坛友微信交流群|倒序 |AI写论文

100论坛币

如题我在做回归的时候选择了固定效应和随机效应但是结果完全一样请问各位老师这个是什么原因啊
STATA命令如下：
xtreg lren_realgdp1 edu pro_fixed open pro_gov road i.year if year<2005 ,fe
xtreg lren_realgdp1 edu pro_fixed open pro_gov road i.id i.year if year<2005 ,re
两个命令得出来的结果完全一样的，请问下这个是什么原因啊

最佳答案

ruyixue 查看完整内容

Please check this file for your reference: https://www.princeton.edu/~otorres/Panel101.pdf There are two reasons for the same results: 1. Use fixed-effects (FE) whenever you are only interested in analyzing the impact of variables that vary over time. Therefore, if there is no impact of time-variant variables, FE and RE are same. 2. You can use the same results to justify your method selection.

分享0 收藏1 回帖

关键词：固定效应随机效应怎么办

回帖推荐

ruyixue 发表于2楼查看完整内容

本帖被以下文库推荐

· 杂七杂八|主题: 471, 订阅: 4

使用道具举报

沙发

ruyixue

发表于 2019-3-5 05:29:05 |只看作者 |坛友微信交流群

Please check this file for your reference: https://www.princeton.edu/~otorres/Panel101.pdf
There are two reasons for the same results:
1. Use fixed-effects (FE) whenever you are only interested in analyzing the impact of variables that vary over time. Therefore, if there is no impact of time-variant variables, FE and RE are same.
2. You can use the same results to justify your method selection.

使用道具举报