Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Yuval Arbel <yuval.arbel@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Thanks and Amazement |
Date | Wed, 23 Nov 2011 12:15:01 -0800 |
Dear Statalist Participants, I personally use Stata during a period of 13 years (starting from 1997), but only recently I became a member of Statalist. I'm also working with several other statistical packages, mainly E-views and Matlab (rarely with SPSS) - and sometimes i like to make comparisons among outputs of different statistical packages. I agree with most of the things said above, but I would like to summarize what are (at least in my opinion) the advantages and disadvantages of Stata. I believe that the hugh advantage of Stata is that it runs like what you might call an open access code. From this reason Stata is the most wonderful statistical package particularly for academic researchers, who look for all kinds of extraordinary new and "sexy" toolboxes. In this field there is no match for Stata anywhere. Just a few examples are procedures like -boxcox-, -stcox-,-heckman-,-selmlog-,-spreg-, bootsrap, competing risk models and unbalanced panels: I'm not familiar with any other statistical packages who can deal with unbalanced panel and competing risk models. In fact, finance scholars, who published the first studies on mortgage default and prepayment while employing competing-risk model, actually programmed the commands by themselves while using FORTRAN. The same thing applies to spatial analysis of auto correlation. On the other hand, I believe that Stata is not a very friendly statistical package for beginners and fresh students with limited needs of simple regressions. The problem is that the basic approach in Stata is based on commands rather than windows (albeit that starting from Stata 6 - there is also the windows options). I find E-Views to be much more friendly in this matter. Another problem is the graph design - which is much nicer with E-View. In fact, when I had to deal with time-series analysis and VAR models, I preferred to work with E-Views where the correlograms look nicer, and it was much more convenient for me to draw the impulse-response function. In sum,I believe that Stata is wonderful in analysing panel data, probit and heckman and special and extraordinary procedures, but its drawbacks lies in time-series analysis On Wed, Nov 23, 2011 at 10:22 AM, Ronan Conroy <rconroy@rcsi.ie> wrote: > On 2011 Samh 23, at 16:16, William Gould, StataCorp LP wrote: > >> Smart people can ask stupid questions and, when they do, it is >> the question that is stupid, not the questioner. > > My wall has a Pearls Before Swine cartoon: > > PIG: I have a stupid question. > RAT: There's no such thing as a stupid question. Just stupid people asking questions. > PIG: Is that me? > RAT: Don't be stupid. > > > r > * > * 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/ > -- Dr. Yuval Arbel School of Business Carmel Academic Center 4 Shaar Palmer Street, Haifa, Israel e-mail: yuval.arbel@gmail.com * * 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/