I think at this point when we cross paths I will be picking up your
tab for the night!
Tom
On Thu, Apr 16, 2009 at 2:40 PM, Nick Cox <[email protected]> wrote:
> Thanks. You would be better off going to a users' meeting, or Stata conference, and buying me a drink if I am there too and you think you owe it to me.
>
> Nick
> [email protected]
>
> Thomas Jacobs
>
> Nick,
>
> As always thanks for the detailed reply. $10 is cheap compared to the
> time I have spent already. Where do I mail the check?
>
> Tom
>
> On Thu, Apr 16, 2009 at 2:26 PM, Nick Cox <[email protected]> wrote:
>> The meta-reply is to rummage in [U] as these matters are explained
>> carefully and fully there. The kind of programming you are doing is
>> fiddly and you need a decent source by your hand and preferably
>> transferred to your brain.
>>
>> In essence, you need compound double quotes `" "' (hereafter c.d.q.)
>> whenever you are nesting double quotes " " as the latter are ambiguous.
>> But c.d.q. do no harm if they are not needed.
>>
>> An example of the problem is "a"b"c". Is this
>>
>> "a"
>>
>> followed by
>>
>> b
>>
>> followed by
>>
>> "c"
>>
>> Or is it
>>
>> "b"
>>
>> nested within a string that starts "a and ends c". You may know what you
>> mean but Stata hasn't a chance because a beginning " and an ending " are
>> of course exactly the same symbol. However c.d.q. are clearly different
>> as `" and "' are distinguishable by programmer and machine alike. Each
>> c.d.q. is logically a single symbol although typed as two (same story as
>>>=, <=, !=, etc.).
>>
>> In essence too: if you mean $ to indicate anything other than a global
>> macro symbol you must escape it with \. That's not an absolute as Stata
>> manages to be a bit smart about e.g. $10 which is what this email would
>> cost you at a decent consultancy rate.
>>
>> Nick
>> [email protected]
>>
>>
>> -----Original Message-----
>> From: [email protected]
>> [mailto:[email protected]] On Behalf Of Thomas Jacobs
>> Sent: 16 April 2009 20:14
>> To: StataList
>> Subject: st: A couple of file command questions for writing a .tex
>> program
>>
>> Fellow listers,
>>
>> I am trying to learn to write Latex files from Stata/Mata with the
>> file commands and have a couple of questions:
>>
>> 1. What is the difference between using simple quotes " xxx "
>> compared with `" xxx "' type quotes? The user manual displays simple
>> quotes for write commands
>>
>> file write handle "first part is"
>>
>> while in the Gini/Pasquini' Automatic Generation of Document article
>> from Stata Journal vol 6 number 1, the typical example
>> uses a format like,
>>
>> file write handle `"first part is"'
>>
>> I would like to know how they are treated differently, if at all.
>>
>> 2. A more complicated question. How does one use a $ sign in a file
>> write command and not have Stata treat it as a global macro? For
>> instance, the end of the following command:
>>
>> file write tex "\begin{tabular}{lrrrrrrrrrrr}\hline\\ \scriptsize
>> Model & `: word 1 of `:colname e(b)'' & p-val & "
>> file write tex "`: word 2 of `:colname e(b)'' & p-val & `: word 3 of
>> `:colname e(b)'' & p-val & `: word 4 of `:colname e(b)'' & "
>> file write tex "p-val & `: word 5 of `:colname e(b)'' & p-val & adj.
>> $R^2$ \\" _n
>> file write tex "\vspace{0.05in}" _n
>>
>> generates:
>>
>> \begin{tabular}{lrrrrrrrrrrr}\hline\\ \scriptsize Model & FFMkt &
>> p-val & SMB & p-val & HML & p-val & UMD & p-val & _cons & p-val & adj.
>> ^2$ \\
>>
>> and I end up with ^2$ as opposed to the $R^2$ term I need for a Latex
>> math string. I also tried
>>
>> `=char(35)'R^2$
>>
>> without any change in results. Any suggestions would be much
>> appreciated.
>
> *
> * 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/
>
--
Thomas Jacobs
*
* 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/
