Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

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

# st: Estimation of Limited depandat variable model with MLE

 From nadir virk To statalist@hsphsun2.harvard.edu Subject st: Estimation of Limited depandat variable model with MLE Date Mon, 23 May 2011 16:17:08 +0300

```Hi All,
I want to estimate the transaction cost measures for cost of selling
and cost of buying in equity markets as is done in
Lesmond, D., Ogden, J, Trzcinka, C.,1999, A new estimate of
transaction costs, Review of financial studies,12,1113-1141.
They have explained the procedure in Appendix B if somebody coouldn't
understand what I am narrating here,
They specified three regions in the distribution of y(i's) and x(i's)
[its a single factor capm model]. They maximized the liklihood
function  with respect to four unknowns (alpha1,alpha2, market beta
and sigma of normal distribution).
The maximization is done  for three regions  simultaneously such that
1-where x's (independant variable) is greater than zero (region1) with
unknown parametrs (alpha1, beta and sigma)
2-where x's are less than zero (region2) with parameters alpha2,beta and sigma
3- when x's are exactly equal to zero.
They maximized these regions simultaneosly and estimated the unknown
paramters. I can't figure it out how to specify it under tobit
command, which fits the model over the whole distribution (if is
possible then excuse me as I don't know how!). But in this case we
need to estimate the alpha1 and alpha2 from region1 and region2
respectively and also the a constant sigma of error distribution.
Any help will be highly appreciated.
Thanks,
Nader
*
*   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 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index