作者oz5566 (阿康)
看板Statistics
标题[问题] P值疑问
时间Thu Dec 29 22:31:41 2016
刚刚跑STATA分析
发现P值好像太大了
Source | SS df MS Number of obs = 25
-------------+------------------------------ F( 4, 20) = 92.31
Model | 479826275 4 119956569 Prob > F = 0.0000
Residual | 25988719.4 20 1299435.97 R-squared = 0.9486
-------------+------------------------------ Adj R-squared = 0.9383
Total | 505814994 24 21075624.8 Root MSE = 1139.9
------------------------------------------------------------------------------
w | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mw | .762513 .646691 1.18 0.252 -.5864608 2.111487
u | 1162.119 307.3389 3.78 0.001 521.0208 1803.216
cpi | 194.0782 176.4702 1.10 0.284 -174.0322 562.1886
gr | 33.77658 93.83759 0.36 0.723 -161.9652 229.5184
_cons | -1173.133 5815.637 -0.20 0.842 -13304.34 10958.07
------------------------------------------------------------------------------
几乎都是大於0.05
请问这是发生什麽问题呢?
这样还能分析吗
--
※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 150.117.206.187
※ 文章网址: https://webptt.com/cn.aspx?n=bbs/Statistics/M.1483021904.A.DDB.html
1F:→ andrew43: 你有看到Adj R-squared = 0.9383吗? 12/29 23:30
2F:→ celestialgod: 你系数的SE都很大 要不要检验一下共线性问题 12/29 23:35
3F:→ celestialgod: 感觉起来 u 跟 _cons的相关系数应该满大的~~~ 12/29 23:39
4F:→ celestialgod: 建议先看看input的correlation matrix 然後确认VIF 12/29 23:40
5F:→ oz5566: 我新手 第一次做这个 我去查名词看看 12/29 23:45