st: Re: RE: fastgini -- not so fast

["Zurab Sajaia" <zsajaia@hotmail.com>]

Thanks alot to Kit and Nick for clarifying the source of problem. I've to 
admit I keep forgetting to put the version statement assuming everybody 
should have the Stata 9 ...

As for Forouzan's other question about the grouped data. When 
option -bin(#)- is used with -fastgini- it first aggregates data into # 
number of groups and calculates Gini on that group level, that's what we 
refer to as `approximate calculation' in the help. Both, Gini itself and jk 
standard errors will differ from the ones calculated using the original, 
ungrouped data but my claim was that for large number of original 
observations (3-4M) and around 100,000 bins, these approximations should be 
quite close to the exact ones while being much faster (thus the name 
'fastgini').

Zurab


----- Original Message ----- 
From: "Nick Cox" <n.j.cox@durham.ac.uk>
To: <statalist@hsphsun2.harvard.edu>
Sent: Saturday, February 17, 2007 9:46 AM
Subject: st: RE: fastgini -- not so fast



> Just to spell out:
>
> 1. The answer to Forouzan is that you cannot use
> this program without upgrading to Stata 9.
>
> 2. A principle for everybody, from the Statalist FAQ,
> is "The current version of Stata is 9. Please specify
> if you are using an earlier version; otherwise, the
> answer to your question is likely to refer to commands
> or features unavailable to you."
>
> Nick
> n.j.cox@durham.ac.uk
>
> Kit Baum
>
> > Fourozan said
> >
> > Dear all,
> > Regarding "fastgini" that was mentioned by Zurab,
> >
> > What does this error mean and how can I fix it?
> > "
> > fastgini income [w=weight], jk
> > unrecognized command:  mata
> > (error occurred while loading fastgini.ado)"
> >
> > It means that you are running Stata 8.x, not Stata 9.x.
> >
> > Against my better judgement I accepted fastgini for SSC without
> > demanding a version statement, as I rationalized that as the routine
> > invokes Mata it was obvious that it needed version 9. But Stata
> > doesn't know that until you try to invoke Mata. I have asked Zurab to
> > submit a new copy of fastgini with the appropriate version statement.
> >
> > Routines submitted to the SSC Archive should always contain version
> > statements in the code, as that information is highlighted in ssc
> > whatsnew to indicate whether a user can make use of a particular
> > package.
> *
> *   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/
*
*   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/