On models in Economics

Chapter 7 is about models, and while the word theory/theoretical is used rather sparsely, the word model/modelling is used about 10 times more often. Here I also start with the meaning of models according to some advanced dictionaries and with an investigation of how the words are used in the AER-articles. Two main interpretations are mathematical models that are studied in “theoretical” articles and empirical models that is the base for statistical estimation. The focus in the rest of the chapter is on the first type of model. It is argued that a Bayesian framework can be used to evaluate the empirical value of mathematical models: The result in the model is then seen as “new information” and the questions is how much it is rational to then change the probability of specific empirical assertions. Realism in assumptions, robustness of the results and lack of competing hypotheses are then important aspects, besides the mechanism being supported by “common sense”.