Two Econometric Paths for Information Recovery in Economic Behavioral Systems

28 Pages Posted: 10 Mar 2015

See all articles by George Judge

George Judge

University of California, Berkeley - Department of Agricultural & Resource Economics

Date Written: March 9, 2015

Abstract

In this paper we assess two economic-econometric information based approaches-paths to recovering behavior related choice information and making inferences from observational data. The traditional path, starts by assuming a stochastic model based on economic, econometric and inferential statistics foundations. The unknown and unobservable parameters of the assumed structured-stochastic model are estimated from a relevant sample of observed data and used for inference and prediction. The other approach, which we emphasize in this paper and call nontraditional, recognizes that our knowledge regarding the underlying behavioral system and observed data process is complex, partial and incomplete, uses a self- organized agent-based algorithmic-representation system, such as networks and machine learning. As a status measure-optimizing criterion we recognize the connection between adaptive intelligent behavior and causal entropy maximization and for information recovery we use an information theoretic basis for estimation, inference model evaluation and prediction. As examples for information recovery and predicting agent’s choices, we cast the problem in the form of an ordered-directed binary and weighted-choice networks and use information theoretic methods to recover estimates of the unknown behavioral parameters. Our objective is to recover expected values across the ensemble that can be computed analytically, without explicitly sampling the configuration space.

Suggested Citation

Judge, George G., Two Econometric Paths for Information Recovery in Economic Behavioral Systems (March 9, 2015). Available at SSRN: https://ssrn.com/abstract=2575834 or http://dx.doi.org/10.2139/ssrn.2575834

George G. Judge (Contact Author)

University of California, Berkeley - Department of Agricultural & Resource Economics ( email )

207 Giannini Hall
University of California
Berkeley, CA 94720
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
43
Abstract Views
361
PlumX Metrics