Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: Predictions and psuedo Rsq in zoib


From   Prerna S <maruiprerna@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Predictions and psuedo Rsq in zoib
Date   Sat, 24 Sep 2011 07:38:35 -0400

Hello Statalisters,

I have estimated the following equation:

zoib y1 x1 x2, zeroinflate (x1 x2) cluster (z) nolog
zoib y2 x1 x2 zeroinflate (x1 x2) cluster (z) nolog

where yi = income from various sources and z = geographical location
(city dummies).

I would like to calculate the pseudo-R squares.

Thus I specified the following:

predict y1_pred

Stata responded with:

(option pr assumed)

I think I am okay with this because I just want to get a sense of the
global measure of explained variance as Ferrari and Cribari-Neto (
2004) suggest for beta regression. I'm assuming that a similar method
on zoib is not so bad.

However, on running a correlation, I get a correlation of -0.5. Why
would the correlation be negative!?! I used the command below.

correlate y1 y1_pred

On the other hand,

correlate y2 y2_pred

gives me a positive correlation.

Am I misinterpreting the result or is this an incorrect way to
ascertain Rsquares for an inflated beta regression?

I am using Stata 11.2. I installed zoib from ssc.

Ferrai, S., & Cribari-Neto, F. (2004). Beta Regression for Modelling
Rates and Proportions. Jl. of Applied Statistics. Vol. 31, No. 7,
799-815.

Thanks.

Prerna
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index