As the very old joke goes, the great thing
about standards is that there are so many to
choose from. What we're seeing here, I suggest, is some
difference across the spread of users in
tribal habits.
Another more recent jest asserts that attaching
asterisks to classes of P-values is `a symbolism
more appropriate to an hotel guide-book than
a serious scientific paper' (P. Sprent JRSS A
133: 139-165, 1970; quotation on p.143). Such
disapproval hasn't stopped it becoming
standard in some fields, and more importantly
in some journals.
Most bizarre of all is that the practice was introduced
by Frank Yates accidentally in his 1937 monograph
on experimental design. The asterisks were used
to indicate successive footnotes, but someone
must have thought this a new standard notation. That's
my version of the history any way: others may
have older references.
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Roger Newson
> Sent: 11 May 2004 17:37
> To: [email protected]
> Subject: Re: st: Stata Estimates to LaTeX Tables
>
>
> At 09:02 11/05/04 -0700, Fredrik Wallenberg wrote:
> >I went back and looked at your enduser document but must
> admit I don't see
> >anything there that is useful to me. It doesn't look to me
> that I can use
> >-listtex- to produce the tables I'm used to (basically the
> ones produced
> >by outtex) where you get the significance mark next to the
> estimate and
> >the standard errors in parenthesis below.
> >
> >If I'm missing anything obvious, let me know.
>
> You should be able to produce tables like you are used to by using
> -parmest- and -dsconcat-, followed by -tostring- to convert
> the estimates
> and SEs to string, followed by -replace- to add the
> parentheses to the SEs
> and the significance marks to the estimates, followed by
> -reshape, long- to
> put the parenthesised SEs below the estimates. The resulting
> resultsset can
> then be output using -listtex-. However, I myself have never
> done this,
> because, in the medical sector, most people prefer estimates with
> confidence limits, possibly followed by left-justified P-values to 2
> significant figures (Stata format %-8.2g) if there is enough space.
>
*
* 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/
