[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: re: logistic function

From   "Nick Cox" <>
To   <>
Subject   st: RE: re: logistic function
Date   Thu, 18 Jun 2009 23:49:21 +0100

I agree, naturally. But my only point is that for anyone who is thinking
"I need to invert the logistic", it is one natural Stata answer. 


Kit Baum

Nick pointed out invlogit(), which does not appear in the function  
list of density functions, but of math functions. Quite so, but it  
doesn't save much typing:

local cum1  1/(1+exp(-((_b[_cons]+ _b[trunk]*mut + _b[mpg]*mum +  
_b[d]*1) - mu)/s))

scalar onea = invlogit((_b[_cons]+ _b[trunk]*mut + _b[mpg]*mum +  
_b[d]*1 - mu)/s)

will indeed generate the same values, but you still have to compute mu  
and s and include them in the function argument.

In my prior example I had the line

gen d = (runiform()>0.7)

which did not appear in the posting.

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index