Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: evaluating probabilities using kdensity


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: evaluating probabilities using kdensity
Date   Thu, 28 Feb 2008 10:01:23 +0000 (GMT)

There is a long tradition of estimating the complement of
what you want ( Pr(x>=525) instead of Pr(x<525) ) in 
survival analysis, where they are sometimes called the 
Kaplan-Meier estimates of the survivor function. This,
however, does not use any smoothing. In essence it uses
the empirical distribution, which is always discrete. 
In Stata you can use -sts list- and -sts graph- to get
these estimates. I would strongly recommend going with
these estimates unless you have very very good arguments
to do otherwise. This is a tried and test technology.

There is some devolpments in terms of the inverse of 
what you want (What value of x has 37% of the 
observations below) which does use smooting, see 
-ssc desc hdquantile- by Nick Cox, and the references
in the helpfile of -hdquantile-.

Hope this helps,
Maarten

--- Conner Mullally <mullally@primal.ucdavis.edu> wrote:
> I have a puzzle that I cannot figure out.  I have a data
> series that runs from 600 or so up to 2200.  What I would
> like to do is estimate a kernel density using this series,
> and then evaluate the cumulative probabilities of different
> values within the range cited above.  For example, I might
> have X=(500, 623, and 2300), which I use to estimate the
> kernel density.  But then I would like to know the
> probability that x<525, as predicted by the results of the
> kernel estimation.  I tried creating a new variable
> consisting of integers between 500 and 2200 and using the
> "at" option with this variable, but I think this changes the
> number of points used to estimate the kernel, which is
> something I do not want.  Thanks in advance for the help!  


-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      ___________________________________________________________
Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
now.
http://uk.answers.yahoo.com/
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index