Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: test for correlation coefficient


From   BISSERY Alvine <[email protected]>
To   [email protected]
Subject   st: test for correlation coefficient
Date   Thu, 18 Sep 2003 17:15:28 +0200

Dear all, 

Is there a test to compare 2 correlation coefficients after regression in 2
populations. I would like to test if they differ.

Thank you

Alvine

-----Message d'origine-----
De: Meissner Chris [mailto:[email protected]]
Date: jeudi 18 septembre 2003 16:52
�: [email protected]
Objet: st: Fwd: RE: nlogitrum


Hi,

I have been trying to implement the nlogitrum command
but I am having no luck!!!

My data involve two upper level nests (fixa, nofixa).
Nofixa is a degenerate nest. fixa has two choices
within the branch: 1 or 2 (dollar or dem).

The structure of my tree given by stata is 

regimedega regimea 
--------------------------
nofixa 0 
fixa 1 
2


This tree was obtained by doing:

nlogitgen regimedega = regimea(nofixa: 0, fixa: 1 | 2)

I then run this:

nlogitrum choice dollar dem opendollar opendem if
year==1995,
group(ctryyr) nests(regimea regimedega)
ivc(nofixa=3.14)



I get the following error message:

could not calculate numerical derivatives
flat or discontinuous region encountered
r(430);


A few other things: 

1. This is not because of the year on which I am
runing the data.

The data look basically like this for a typical
country in 1995


ctryyr choice regimea dollar dem opendollar opendem 
open

| 676305 1 0 0 0 0 0 71.85 |
| 676305 0 1 1 0 71.85 0 71.85 |
| 676305 0 2 0 1 0 71.85 71.85 |

+-----------------------------------------------------

2. I get the same error message no matter which
arbitrary constraint I
put in for the inclusive value (ivc). 

3. Finally and maybe most importantly: Is my data
structured
appropriately? There is a cryptic message (p.247)
about data structure
in Heiss' (excellent) Stata Journal article (vol2
Number 3 pp. 227-252)
about data setup that says, "nlogitrum only allows
explanatory variables
to directly enter the conditional probabilities of the
alternatives...it
does not make a difference at all if the nest-specific
variable is
specified for a nest or for all alternatives within
the nest..." I have
no idea what the implication of this is for my data.
As I understand it,
it only applies to variables we think might affect the
choice of nests
but not subalternatives in a nest.

Anyhow, any help or support anyone can offer would be
greatly
appreciated. 

Chris Meissner
University of Cambridge


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design
software
http://sitebuilder.yahoo.com


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index