Por que usar o Mathematica em aplicações de Economia e Econometria?
Lendo o texto "Mathematica as an Enviroment for Doing Economics and Econometrics", de David Beasley, do Boston College, me deparei com este trecho que vale a pena reproduzir.
É útil para quem quer trabalhar com economia computacional e está em dúvida sobre qual software (ou linguagem de programação) escolher para investir seu tempo de estudo.
"Numerous environments exist that will be of interest to graduate students in economics. These include such packages as GAUSS, MATLAB, GAMS, Mathematica, Maple, Derive, Reduce, Macsyma, RATS, TROLL, PCGIVE, Ox, SAS, SPSS, S, S+, and Stata. While each of these is potentially useful, they are not all the same type of environ. At the risk of oversimplifying, the rst two, GAUSS and MATLAB are quite specially designed programming environments for doing matrix-type calculations easily and efectively.
GAMS is also a special-purpose environment; it is designed for building and manipulating large-scale economic models. Similarly, the last nine are basically statistical packages, allowing one to do standard, and not so standard, statistical calculations with relative ease and exibility. Among these, RATS, TROLL, and PCGIVE are more econometrically oriented, and S+ is oriented towards data analysis. Stata is somewhere between a statistical package and a fully extensible econometrics programming language. The remaining products, particularly Mathematica, Macsyma, and Maple, are more general environments that allow for a wide spectrum of operations, encompassing most of what has been mentioned, albeit perhaps at the cost of some e ciency in carrying out speci c tasks. Among these, Mathematica and Maple are arguably the two most popular and widely used.
Thus, the strategy I suggest would point toward selecting an environment like Mathematica or Maple.
These provide powerfulmodeling, graphical, statistical, and programming facilities that allow the user, within a single environment, to do almost everything. And even if more special-purpose routines are eventually called for, they allow for rapid prototyping and testing of projects that makes subsequent implementation in other environments very much easier and surer.
Both Mathematica and Maple, for very good reasons, have their devoted adherents. I happen to be a Mathematica sort, and it is Mathematica that I want to deal with here. It is not, however, my purpose to "sell" Mathematica. Rather I want to try to elucidate for those who are considering computational environments what Mathematica's pros and cons are for use in economic research. To this end, I will try to explain some of Mathematica's salient features and illustrate the kinds of economic research that have successfully been accomplished using Mathematica."