Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Problem with estout and rename


From   Cory Smith <corybsmith@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Problem with estout and rename
Date   Wed, 20 Mar 2013 12:31:09 -0400

Thanks for sending the piece along--it's definitely interesting and
something to consider. I will say that in our many models only one
coefficient really matters, but perhaps there would be a nice
graphical way to display the results.

Cory

On Wed, Mar 20, 2013 at 12:10 PM, Nick Cox <njcoxstata@gmail.com> wrote:
> Much in this territory depends on tribal habits and traditional
> rituals and the usual advice in such matters is for outsiders to keep
> quiet and adopt a benign neutral expression.
>
> However, it strikes me that with many models and many predictors,
> showing which predictors are included, or significant at some level,
> or otherwise notable, calls for a graph as well as a table. I did say
> "as well as", although sometimes I would go further.
>
> Indeed this is an old idea.
>
> stat.columbia.edu/~gelman/research/published/dodhia.pdf
>
> is one accessible argument in this vein.
>
> Nick
>
> On Wed, Mar 20, 2013 at 3:37 PM, Cory Smith <corybsmith@gmail.com> wrote:
>> Hi Rebecca,
>>
>> Thanks for your concern about the clarity of the piece. In general, I
>> think the merge aspect of the rename() function comes in handy when
>> there are different ways to measure or proxy for the same variable. To
>> give a crude simplification in my case, I am looking at how disease in
>> location A predicts disease in location B. But, since it takes time
>> for disease to spread across space, maybe I should look at the disease
>> in A from one month ago (i.e. a lag).
>>
>> In a big table with 8+ regressions, it will be easier for a reader
>> just to scan one line to determine how many models show such a
>> relationship (rather than having to bounce across two lines depending
>> on whether there is a lag or not). Other lines in the table and the
>> text in the paper will make clear which measures are used in a given
>> regression.
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index