Statalist


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

Re: st: Can STATA be used for interactive webb-based analyses


From   Sergiy Radyakin <serjradyakin@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Can STATA be used for interactive webb-based analyses
Date   Fri, 16 Oct 2009 13:43:43 -0400

Dear Roland,

1) there is a technical possibility of running Stata driven by queries
submitted from the webpages.

2) there are technical difficulties in terms of how exactly you want
to organize this (otherwise your page will soon become unresponsive,
e.g. after -shell format c:- submitted by some curious 5y.o.),
specifically security issues, data transfer (you don't want to let
users send you 300GB of their census data, right?) etc

3) license issues. (I'd like Stata to be more specific about what is
allowed and what is not here) Bill Gould said in the DC meeting that
they want to avoid the "extreme form of competition", where Stata
competes with itself (where one user buys one copy of Stata and lets
everyone in the world use it over the Internet). It might be clearly
hurtful for the business. On the other hand, some econometric packages
have done exactly this a long time ago. SHAZAM is a UBC-based
(University of British Columbia, where I studied a while ago)
econometrics package, and "Run Shazam over Internet" is a prominent
feature on its homepage:

http://shazam.econ.ubc.ca/
http://econometrics.com/
http://shazam.econ.ubc.ca/runshazam/

4) In your particular case you don't need Stata, you just write a
small VB-script within the page, that computes
y_hat=SUM_i(X_i*Beta_hat_i)
where the betas are coming from your model estimated in Stata in advance.

Best regards, Sergiy Radyakin



On Fri, Oct 16, 2009 at 4:17 AM, roland andersson
<rolandersson@gmail.com> wrote:
> I wonder if STATA can be used to program interactive webb-based
> analyses of large databases. We have a large national quality-database
> for gallbladder surgery which is continously updated. The aim would be
> to let patients find out the probability of adverse outcomes in
> relation to age, sex, hospital volume, previous health
> characteristics, current disease etc. They could than choose for
> themselves if they want to have surgery or not.
>
> Roland Andersson
> *
> *   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/
>

*
*   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/



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