Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: FW: ML for logit/ologit

From   Yuval Arbel <>
Subject   Re: st: FW: ML for logit/ologit
Date   Wed, 16 Nov 2011 12:58:42 +0200


Obviously, my advice is based on my own subjective experience. I
personally finds the work in MATLAB more convenient in the cases that
I have to define non-linear functions - and I even published a paper
with non-linear procedures I ran on MATLAB. I believe that the wisdom
is to exploit the specific advantages of each statistical package.
There are many other cases where I prefer working with STATA.

Moreover, if the bug is not a syntax error - I believe this is a very
serious bug - because it means the program does not know to identify
the very basic sign of power

On Wed, Nov 16, 2011 at 12:27 PM, Jennyfer Wolf
<> wrote:
> Dear all,
> I am working with the nl-function and with the logistic function model log 4.
> Could somebody please explain me why I do not get the same outout if I type:
> 1. nl log4: VAR1 VAR2 if VAR3=="xxx"
> and
> 2. nl(VAR1={b0}+{b1}/(1+exp(-{b2}*(VAR2-{b3}))) if VAR3=="xxx")
> Actually 1. should just be an abbreviation of 2.?
> If I run command 1, I get estimates for b0, b1, b2, b3, if I run
> command 2 I only get an estimate for b0 and b1-b3 are 0.
> Could somebody explain how to write the formula "in full" (like in
> 2.), to get the same results as in the first formula.
> Thank you very much!
> Jennyfer
> *
> *   For searches and help try:
> *
> *
> *

Dr. Yuval Arbel
School of Business
Carmel Academic Center
4 Shaar Palmer Street, Haifa, Israel

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index