Mathematica as an Environment for Doing Economics and Econometrics

Introduction Sooner or later, most of today's graduate students in economics will be faced with a critically important decision: which computational environment to learn and use. And more often than not, they will be ill equipped to make this choice. While various of a student's professors and classmates will undoubtedly advocate speci c software, such suggestions are often salted with confusing contradictions and unassessable evangelism, and, to make matters worse, local computer facilities will likely make only certain of the alternatives available at reasonable cost. The need for this choice may possibly come to some as an unpleasant and belated surprise. Initially, course emphasis is likely placed on theory. But at some point, the rudeness of \doing" arises|sometimes not until well into the graduate curriculum, and sometimes not until the beginning of thesis writing. In years past, this problemmight have been resolved simply by the student's learning to use a canned regression, statistical, or econometric package. But, more and more, students are nding the need to learn the programming skills to develop one's own models and statistical tools. And thus the daunting situation arises in which the student must undertake a major new learning experience totally distracting to any intended economic research. Students' reactions to this vary. Some take time out to determine what software environment is needed and learn to use it well before returning to their economic researches. Some do the same, but get lost in the lure of their new-found game, never to return to economic research again. Others try to learn a software environment and simultaneously continue researches. Yet others compromise, mechanically learning \just what's needed" of particular software packages to keep their research momentum alive. A few simply