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: how to generate mean for a secquence of five year of a variable
From
Nick Cox <[email protected]>
To
"'[email protected]'" <[email protected]>
Subject
RE: st: how to generate mean for a secquence of five year of a variable
Date
Thu, 5 May 2011 15:48:11 +0100
As I understand, that reduces the data. You'd need to -expand- and -merge- back.
For blocks, you just round the time variable and fire up -egen-'s -mean()-. I just gave an example in another thread.
Nick
[email protected]
Oliver Jones
I have never used -rolling- and just had a quick look at the help
file of it.
Couldn't -rolling- be used for both possibilities, i.e. calculating
the mean for separate blocks of five years and moving blocks, by
specifying the stepsize(5) option?
Am 05.05.2011 16:23, schrieb Nick Cox:
> If that's the right diagnosis:
>
> -rolling-
>
> -mvsumm- (SSC)
Oliver Jones
> ok, now I got it. You want some kind of running-mean.
> The function -lowess- has an option to create a new
> variable that is a running mean.
> BUT, this is not exactly what we want here, because this
> mean is calculated at the center of the running window.
> We want the mean at the right end of the window.
>
> Am 05.05.2011 15:53, schrieb prakash singh:
>
>> yearid countryid var1 var2
>> 1980 1 7.25
>> 1981 1 8.25
>> 1982 1 9.5
>> 1983 1 7.5
>> 1984 1 7.75 8.05 (mean of first five observation of var1)
>> ...........................................
>> ........................................
>> 2001 1 8.25
>> 2002 1 9
>> 2003 1 9.5
>> 2004 1 8
>> 2005 1 10 8.95 (mean of last five observation of var1)
>>
>> similarly for all other country.
*
* 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/