 Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

st: RE: Rankings according to percentiles

 From Nick Cox To "'statalist@hsphsun2.harvard.edu'" Subject st: RE: Rankings according to percentiles Date Wed, 30 May 2012 17:34:31 +0100

```Why struggle when Stata can -search-?

. search percentile rank

[...]

FAQ     . . . . . . . . . . Calculating percentile ranks or plotting positions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  N. J. Cox
7/05    How can I calculate percentile ranks?
How can I calculate plotting positions?
http://www.stata.com/support/faqs/stat/pcrank.html

[...]

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

Fabian Schönenberger

I have a panel data set consisting of companies (id) and different fundamental variables (for instance leverage ratio) for different point in times (t). In order to analyse how certain factors correspond with others (for instance leverage ratio with cash reserves), I would like to assign to each observation (leverage ratio at a specific point in time) the corresponding percentile rank. Afterwards, I can analyse if with increasing leverage ratio (percentile ranks) the cash ratios increase, too. With this approach I avoid errors-in-variables bias.
One way would be to take the percentile ranks from the command summarize and assign with if formulas each observation to the rank. Are there other more direct ways to do that?

*
*   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/
```