Bookmark and Share

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: Reversed coefficient sign for exponential MLE vs. ereg


From   Gordon Burtch <gburtch@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Reversed coefficient sign for exponential MLE vs. ereg
Date   Fri, 27 Jan 2012 18:13:40 -0500

Hello,

I am trying to implement the ereg (exponential regression) estimator using ML, and am encountering something odd. I have written a .do file as follows...

program define myexponential
	version 10.0
	args lnf x
	local lambda = exp(`x')
	quietly replace `lnf' = ln(`lambda') - `lambda'*$ML_y1
end

Maximizing a constant only model with this likelihood function produces a coefficient estimate that matches with that obtained using the ereg command. However, the sign on the coefficient is reversed for some reason. I noticed as well that the log-likelihood is much different (-765.96 versus -1607.53). Does anyone have an idea as to what might be causing this? If I invert `x' when defining lambda (i.e., lambda = exp(-`x')), then the correct coefficient is obtained (though the difference in LL between ereg and my implementation is still quite different).

Any help is appreciated! Regards,

Gord Burtch


*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index