Correct. Stata has not always done it this way.
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of
> Stas Kolenikov
> Sent: 03 February 2006 20:22
> To: [email protected]
> Subject: Re: st: RE: kdensity with user defined kernel functions
>
>
> That's interesting -- I was just going to write that you can dig into
> kdensity.ado to put in your own kernel... but you cannot, it is indeed
> in the executable. I think it used to be out there in ado files a
> version or two ago, so if you have previous reincarnations of Stata (I
> have my Stata 5 somewhere deep on the CD backups...), you can (i) copy
> kdensity.ado into a new file, say mykdens.ado; (ii) change the program
> name at the top; (iii) put your own kernel; (iv) see Stata crash...
>
> On 2/3/06, Nick Cox <[email protected]> wrote:
> > At worst, I think you need to write everything
> > yourself.
> >
> > For speed (and possibly other reasons) the
> > hard work in -kdensity- is handed to part of
> > the executable and I can't see any hooks for
> > using your own kernel. So cloning -kdensity-
> > would get only part of the way there.
> >
> > Alternatively, Ben Jann's -kdens- package
> > seems more transparent.
> >
> > Nick
> > [email protected]
> >
> > D'Exelle Ben
> >
> > > I need to make a kernel density estimation with a kernel
> > > function that is not supported by the kdensity command. Is
> > > there a way to define your own specific kernel function
> > > within the kdensity command? If this is not possible within
> > > the kdensity command, is there another way to make kernel
> > > estimations with user-defined kernel functions?
*
* 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/
