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

st: Defining observation specific truncation points with truncreg


From   Eric Sevigny <els21+@pitt.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Defining observation specific truncation points with truncreg
Date   Thu, 30 Sep 2004 17:34:09 -0400

Hello,

I'm attempting to estimate a truncated regression model using alternate lower level truncation points, which vary by year in my dataset. The Stata reference manual states one can use either a specific cut value or variable to identify the limit. My question is, can the -truncreg- option -ll(varname)- be used to set observation-specific limits and, if so, how? The reference manual is silent on this. It does refer one to the -tobit- entry in the reference manual, but this is of little help as -tobit- limits do not accept the -ll(varname)- option. I initially thought -ll(varname)- might be coded like -cnreg- (i.e., -censored(varname)-), taking values of -1, 0, 1 for lower limit, in sample, and upper limit indications, but this does not seem to be the case. It also does not seem to be a simple 0,1 coding, although I cannot be sure.

If I may summarize, does -truncreg- support observation-specific truncation points? If so, how exactly would I implement it? In other words, how would I code varname to indicate these alternate truncation points?

Eric Sevigny
University of Pittsburgh
*
* 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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index