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

st: Using -byable- and the -by- commands correctly?


From   "David Winter" <[email protected]>
To   <[email protected]>
Subject   st: Using -byable- and the -by- commands correctly?
Date   Mon, 31 Oct 2005 10:23:59 -0000

Dear Statalisters,

Firstly, thanks to Svend Juul for the helpful comments on survival
analysis.

The following 4 records are part of a 27 record dataset sorted by
ageband (5,10,15,20..85) within sex (MALE,FEMALE).

Sex            Ageband   Observed  Expected  PoissStat 
MALE           5         5         .2585
MALE           10        11        .6054
MALE           15        6         .6516
MALE           20        5         .6435

I have written a program to calculate a two-tailed poisson statistic,
that has three arguements:

`1' = Observed number
`2' = Expected number
`3' = variable to hold the poisson statistic

I have made the program -byable- :

Program poissontt, sortpreserve byable(recall, no header)
..
..
local poissval = 0 /* macro to store the result */
..
local observed = `1'
local expected =`2'
local replacevar = "replace `3' = "
..
<define poissval>
..
`replacevar' `poissval'
end

In a -.do- file, I run the commands

gen PoissStat = .
by sex ageband: poissontt Observed Expected PoissStat

Stata diligently loops through the data and using -set trace- I have
tracked the iterations. It does not appear to pick up the values of
Observed and Expected for each record, rather those of the first record
only.  The calculation is correct for record 1 (Obs=5, Expect=.258,
poisson(two-tailed=.0000153) and the PoissStat variable is updated in
each record, but with the value for record 1 only.

Is there something I am missing either in the program or in using the
-by- command?

With Best Wishes,

Dave

*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index