Statalist


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

st: Box-Cox double hurdle


From   Patrick Hitayezu <phitayezu@yahoo.fr>
To   statalist@hsphsun2.harvard.edu
Subject   st: Box-Cox double hurdle
Date   Mon, 17 Aug 2009 13:50:00 -0700 (PDT)

Dear All,

I am trying to estimate a Box-Cox Double-Hurdle model using the code provided by Moffatt (2005). I have set the value of lambda at 0.7 and the seed at 12678741. I have 17 variables in my model and 514 observations. The dependent variable in my model exhibits a strong positive skewness (to justify the use of the Box-Cox DH). Tobit and Double-Hurdle results (before Box-Cox transformation) are running well with some significant p-values. However, I am encountering problems after Box-Cox transformation as follows:

1. NONE of the p-values in the Box-Cox tobit is turning out to be significant and the Prob > chi2 = 1.0000 !?

2. The Box-Cox p-tobit is reporting this error:

initial:       log likelihood =     -<inf>  (could not be evaluated)
could not find feasible values
r(491);

2. ALL the p-values in eq1 of the Box-Cox Double Hurdle are insignificant and the Prob > chi2 = 1.0000 !?

Please help me find the way out of this problem.

Regards,
Patrick


      

*
*   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   |   What's new   |   Site index