用stata显示变量之间相关系数

pwcorr and corr should give you the same correlation, the sig option for pwcorr is to show the p-value of the correlation.

If you want to run the corr for all the variables in the data, using "pwcorr, sig", if you want for some variables, then list the variables after pwcorr, see "help pwcorr".

diligentsai 发表于 2014-2-9 16:53
首先谢谢您！我试过用pwcorr的命令语句，发现它与corr命令语句所显示的系数不一样，请问两者有什么区别。 ...

pwcorr wage educ,sig star(0.05)

pwcorr

可以使用连玉君老师的命令，pwcorr_a。

9999999999

1、首先下载个pwcorr_a的do文件（这个论坛中已有）
2、最后输入logout, save(文件名) word replace: pwcorr_a 变量名,star1(0.01)star5(0.05)star10(0.1)，即可

harleych 发表于 2014-2-9 05:12
如果变量不是很多的话可以用另一个方式
reg y x1 noconst
estimates store mm1

你好，請看這裡，pwcorr相關性很好，可是回歸后回歸係數卻全都沒有顯著性，能解釋一下嗎？

jjjj6666 发表于 2014-2-9 21:57
pwcorr and corr should give you the same correlation, the sig option for pwcorr is to show the p-val ...

Hello,here's one problem!You can see that there's a good correlation between INR and independent variables,nonetheless,no significant coefficients between INR and these independent variables!WhY?

佼佼者爱发呆 发表于 2016-11-22 10:09
1、首先下载个pwcorr_a的do文件（这个论坛中已有）
2、最后输入logout, save(文件名) word replace: pwcor ...

你好，请问pwcorr_a的do文件在哪呢？谢谢

谢谢分享