-
-
Replying to @eigenrobot
I mean, this was fairly well known and you can see an intuitive understanding of what FE measures in a fair number of the papers using it. But the fix is cool.
1 reply 0 retweets 1 like -
Replying to @nickchk @eigenrobot
gonna run it on the FE paper i'm writing now and see what happens
1 reply 0 retweets 1 like -
Replying to @nickchk @eigenrobot
Update: was surprisingly super easy to apply the fix, which changed nothing hell yeah
1 reply 0 retweets 1 like -
Replying to @nickchk @eigenrobot
It's literally (1) Get within-group SD in treatment, (2) run FE using de-meaning with inverse SD as weights. Done.
1 reply 0 retweets 1 like
Replying to @nickchk
oh this is cool! and intuitive. one more reason to program all of my own estimators I guess
5:59 AM - 7 Feb 2018
0 replies
0 retweets
1 like
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.