Hi, last week I enquired of the list as to the availability of Patrick Royston's package 'ptrend' . Thanks to Nick Cox for retrieving the package, to Kit Baum for placing a copy on SSC and, of course, to Patrick for producing it and making it available.
Many thanks to all.
Alan Kelly
--On Thursday, June 6, 2002 15:03 +0100 Nick Cox <[email protected]> wrote:
Kit: Please put this package up on SSC for Patrick.
ptrend Stata module for trend analysis for proportions
ptrend calculates a chi-square statistic for the trend (regression) of
pvar on xvar, where pvar is the proportion rvar/(rvar+nrvar). A variable
called _prop, containing the values of pvar, is left behind by ptrend.
ptrend also gives a chi-square test for departure from the trend line.
ptrendi does the same analysis on user-entered data; the proportions are
now p1 = r1/(r1+nr1), p2 = r2/(r2+nr2), etc.
____________________________________________________________
Hi,
I'd appreciate assistance in locating a copy of Partick Royston's 'ptrend' ado file.
I know that this was sent to the list back in 1996 and I see that Fred Wolfe kindly replied - with a copy - to a query last year....
ptrend has one simple attraction compared with 'opartchi' - it presents the regression fit simultaneously with the test for trend and as I'm testing upwards of 1,000 different sets of observations, this facility is convenient.
Many thanks,
Alan
Dr. Alan Kelly
Biostatistician
Director, Small Area Health Research Unit (SAHRU)
Department of Community Health & General Practice
Trinity College Dublin
