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

st: RE: historical maximum


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: historical maximum
Date   Tue, 27 Sep 2005 10:19:55 +0100

Kit mentioned panel data. For the record, 
-egen, record()-, which was written by Kit 
and myself, pays no attention to whether 
data have been -tsset-. The handle useful
for panel data is a standard -by()- option. 

As it happens, I guess that this is a nuance
not important in most projects. The record 
returned will be the _known_ maximum (or 
minimum) whenever any observations have missing values
or there are gaps. 

A key point about missing values is that in Stata

max(large known number, missing) 

is always "large known number". This may 
come as a surprise once you have learned 
the hard way that in Stata numeric missing
is treated as higher than any non-missing 
number. But usually it is a feature. What 
do you want 

max(0, 1, 2.71828, 3.14159, 42, .) 

to return? 42 or . ? 

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

Kit Baum
 
> Michael suggested
> 
> You can get the running/cumulative maximum using subscripting, such as
> 
> gen cummax=x in 1
> replace cummax=max(x,cummax[_n-1]) in 2/l
> 
> If you have panel data or missing data/gaps, you may want to check  
> out using
> time series operators  by using tsset.
> 
> 
> The egen function record (think of setting a record...) handles the  
> nitty-gritty details quite nicely. It is available in the egenmore  
> package from ssc. It will deal with panel data and missing values.

*
*   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