Computational Techniques for Modelling Learning in Economics offers a
critical overview of the computational techniques that are frequently
used for modelling learning in economics. It is a collection of papers,
each of which focuses on a different way of modelling learning,
including the techniques of evolutionary algorithms, genetic
programming, neural networks, classifier systems, local interaction
models, least squares learning, Bayesian learning, boundedly rational
models and cognitive learning models. Each paper describes the technique
it uses, gives an example of its applications, and discusses the
advantages and disadvantages of the technique. Hence, the book offers
some guidance in the field of modelling learning in computation
economics. In addition, the material contains state-of-the-art
applications of the learning models in economic contexts such as the
learning of preference, the study of bidding behaviour, the development
of expectations, the analysis of economic growth, the learning in the
repeated prisoner's dilemma, and the changes of cognitive models during
economic transition. The work even includes innovative ways of modelling
learning that are not common in the literature, for example the study of
the decomposition of task or the modelling of cognitive learning.