Teaching Computational Economics to Graduate Students

• Authors: David Andrew Kendrick
• Venue: Computational Economics
• Year: 2007

The teaching of computational economics to graduate students has mostly been in a single course with a focus on algorithms and computer code. The shortcoming with this approach is that it neglects one of the most important aspects of computational economics--namely model development skills. These skills are the ability to conceptualize the science, engineering and economics of a problem and to convert that understanding first to a mathematical model and then to a computational representation in a software system. Thus we recommend that a two course sequence in computational economics be created for graduate students with the first course focusing on model development skills and the second course on algorithms and the speed and accuracy of computer codes. We believe that a model development course is most helpful to graduate students when it introduces the students to a wide variety of computational models created by past generations and ask them to first make small modification in order to better understand the models, the mathematics and the software. This in turn is followed by encouraging them to make more substantial modifications of the students' own choosing so as to move the models in directions that permit the students to address current economic problems. We think that the key element of this process is it's enhancement of the creative abilities of our students.