Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Interpolation using cipolate


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Interpolation using cipolate
Date   Wed, 14 Apr 2004 19:12:26 +0100

-cipolate- is a user-written command on SSC for cubic 
interpolation. As Richard surmises, the logic is that 
you must interpolate gaps in a response with respect 
to a second variable (e.g. time), just as with official 
Stata's -ipolate-, which is for linear interpolation. 

Nick 
n.j.cox@durham.ac.uk 

Richard Williams replied to Joe J. 

> >A naive question. I just insalled cipolate to fill missing 
> values for one 
> >variable. The documentation tells me that interpolation is 
> done using two 
> >points before and two points after the missing value. That 
> said, why is it 
> >that the procedure requires a second variable? In other words, for 
> >replacing missing values in variable y, why we need to 
> include variable x 
> >in the command?
> >
> >. cipolate y x, gen(yprime).

> I don't know about -cipolate-, but the docs for -ipolate- 
> (reference manual 
> G-M, pp. 182-183) say that Y is a function of X.  The value for 
> interpolated Y is computed using values from the surrounding 
> Xs and Ys that 
> are not missing.
> 
> If you are going to interpolate, your data have to be sorted 
> in some way, 
> right?  So having an X accomplishes that.  The distance 
> between Xs will be 
> related to the interpolated differences between Ys.

*
*   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