Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: csipolate not csipolating
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: csipolate not csipolating
Date
Mon, 21 May 2012 19:04:14 +0100
We traced Michael's problem to my bug in -csipolate- in its
over-paranoid treatment of singletons. I'll post to the list when the
fixed version of the code is available on SSC.
Nick
On Sun, May 20, 2012 at 3:10 PM, Michael Mulcahy
<[email protected]> wrote:
> My apologies for omitting the source for the commands -csipolate- and -cipolate- (SSC) - those are user-written commands by Nicholas Cox.
>
>
> Thank you for the offer to take a look at the data - I'm happy to send it, but it's probably not worth your trouble, as I was just looking to compare the cipolate results to the csipolate results.
>
>
> Mike
>
>
>
>
> ----- Original Message -----
> From: Nick Cox <[email protected]>
> To: [email protected]
> Cc:
> Sent: Sunday, May 20, 2012 6:50 AM
> Subject: Re: st: csipolate not csipolating
>
> -cipolate- and -csipolate- (SSC) are user-written commands. Please
> remember that you are asked to explain where user-written commands you
> refer to come from.
>
> That said, all the hard work here in -csipolate- is done by official
> Mata functions.
>
> I don't see any obvious reason why this shouldn't work. You could try
> sending me the data privately to check, but I am going off-line now
> and will not be accessible for some hours.
>
> Nick
>
> On Sun, May 20, 2012 at 2:30 PM, Michael Mulcahy
> <[email protected]> wrote:
>
>> I'm trying to use csipolate to interpolate missing values for 1 year in a panel dataset of city-year observations. None of my cities have observations for 1997, they all have observations on the variable in question for 1992 thru 2002. I'm writing:
>>
>> by id: csipolate dues year if year > 1991, gen(csipdues)
>>
>> The new variable, "ipdues", has missing values for 1997.
>>
>> The cipolate command (with analogous syntax), on the other hand, works. I can just use the "cipdues", but I wanted to compare that to the csipolate result.
>>
>>
>> Am I not understanding some "no singleton values" issue here?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/