Clive Payne
>
> I'm a new Stata user. While I'm very impressed with the
> range of statistical
> models which can be fitted in Stata and its
> programmability, I'm less impressed with the formatting of
> the standard modelling output (although I must admit to
> only having used regress so far). Often there is a spurious
> accuracy with too many significant digits in the output and
> decimal points are not often alligned. So my question is:
>
> is there a GLOBAL way of setting/resetting the number of
> significant digits and the number of digits after the decimal point?
>
> (I could do this in Glim 25 years ago!)
As one of the few geographers (notionally) on this list,
it is a pleasure to welcome an (ex-?)geographer.
The short answer is no.
The broad rationale is presumably
that users can round easily, but they might find it
difficult or impossible to reverse rounding they judge
too vigorous.
You could attach the same -format-
to all numerical variables quite easily, but that would not
be echoed to all commands. What would you like? That
absolutely every number be to 2 decimal places or to
3 decimal places?
-tabstat- with a -format()- option is one command
I like because it offers the possibility of
sensible rounding, although when I specify %3.2f
it is irritating that sample numbers get shown
in the form 123.00, as I have complained to its
author.
Similarly, there are quite a few commands which produce matrix
output, and the -format()- option of -matrix list- then
gives complete control. I don't know anyone who can interpret
correlations to 4 d.p.: alternatives to -correlate- which show 2 or
3 d.p. already exist or can be set up easily.
Nick
[email protected]
*
* 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/
