st: data truncation conditional on some criteria?

[Marcello Pagano <pagano@hsph.harvard.edu>]

Dear Statalists,

Sorry if the msg is sent for the second time.

My insterest is to study the determinants of the total credit line (CLt) 
an credit card holder may apply for. However, the application will only 
be observed when the applicant asks for augmenting her credit line, 
given she has already the card user with some initial credit line. 
Therefore, the RHS variable will be the level of total credit card line 
she applies for (CLt), where CLt is the sum of the initial credit card 
line she has (CLi) and the amount of credit card line she asks to 
augment (CLa).

My problem is truncation happens not directly on CLt, but on CLa, which 
means only when the applicants asks to augment her credit card line 
(therefore CLa>0) the total level of credit card line she applies for 
(CLt) will be observed. The applicants who do not apply for augmenting 
their credit card line are not observed.

I tried -truncreg- in STATA, but apparently the command does not allow 
me to setup any selection criteria. And my data doesn't fit the Heckman 
selection model neither. So will anyone happen to know any other 
suitable model for my case or kindly give any suggestion to me?

Will appreciate very much any advice on this.

Xiaoqiang

*
*   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/