[Help-gsl] Re: Maximizing likelihood functions - GSL vs. Matlab

[AJ Bostian]

OK, so I found the issue, but it baffles me.  I have a space-delimited 
datafile, and the first line of it is

20 0 0.025 0 0.12

and I'm reading it in using an ifstream called "data":

data >> junk >> Y[0] >> X[0][0] >> X[0][1] >> X[0][2] >> X[0][3];

where junk, X[], and Y[][] are double.  But Y[0]=1 (!!!) when I examine 
the data in the debugger.  This is the only data element that this 
happens to, but it happens every time I load this file.  Any thoughts?

After I fixed this, the results were much more comparable with the 
commercial results, usually to within 2 decimal places of precision. I'm 
now trying to figure out the remaining differential.  I'm using the 
gsl_multimin minimizers.

AJB


AJ Bostian wrote:
> I recently programmed up some probit and logit routines in GSL and in 
> Matlab.  (The Matlab machine is an i386 and the GSL machine is an 
> x86_64, both running linux)  I then compared the output from each of 
> these to the output from some commercial statistical packages like SAS 
> and Stata.
>
> The output from the Matlab code matches up with the commercial output 
> exactly -- the parameter estimates, maximizing value of the likelihood 
> function, etc. are identical.  However, the GSL output tends to show a 
> slight improvement in the maximizing value of the likelihood function. 
> The improvement is not large, say a 0.1% difference, but the parameter 
> estimates are noticeably different (5-10%) due to the nonlinearity of 
> the function.
>
> Just to check, I plugged the GSL parameter estimates into the Matlab 
> objective function.  The value of the Matlab objective using the GSL 
> parameter estimates is indeed very slightly worse.
>
> I thought that this might be due to differences in precision, but 
> Matlab/SAS/Stata all seem to do calculations in double, which is what I 
> used.  I also thought that the choice of GSL minimizer might make a 
> difference, but the result holds under all GSL minimizers.
>
> Obviously, I would like to think that the improvement in the maximizing 
> value of the likelihood function using GSL is real, but it's hard to go 
> up against Matlab/SAS/Stata.  Anyone have any thoughts about why this 
> difference might exist, and whether the improvement is legit?
>
> Regards,
> AJ Bostian
>
>
> PS: Can't beat GSL for speed, though.  I blew through 20 iterations of a 
> probit problem in about 0.25 sec using GSL, compared to about 3 sec for 
> Matlab.
>
>
> AJ Bostian
> Dept. of Economics
> University of Virginia
> Rouss Hall 114
> Charlottesville, VA  22904
> Email: address@hidden