In the theory and practice of econometrics the model, the method and the data are all interdependent links in information recovery-estimation and inference. Seldom, however, are the economic and statistical models correctly specified, the data complete or capable of being replicated, the estimation rules ?optimal? and the inferences free of distortion. Faced with these problems, Maximum Entropy Economeirics provides a new basis for learning from economic and statistical models that may be non-regular in the sense that they are ill-posed or underdetermined and the data are partial or…mehr
In the theory and practice of econometrics the model, the method and the data are all interdependent links in information recovery-estimation and inference. Seldom, however, are the economic and statistical models correctly specified, the data complete or capable of being replicated, the estimation rules ?optimal? and the inferences free of distortion. Faced with these problems, Maximum Entropy Economeirics provides a new basis for learning from economic and statistical models that may be non-regular in the sense that they are ill-posed or underdetermined and the data are partial or incomplete. By extending the maximum entropy formalisms used in the physical sciences, the authors present a new set of generalized entropy techniques designed to recover information about economic systems. The authors compare the generalized entropy techniques with the performance of the relevant traditional methods of information recovery and clearly demonstrate theories with applications including * Pure inverse problems that include first order Markov processes, and input-output, multisectoral or SAM models to * Inverse problems with noise that include statistical models subject to ill-conditioning, non-normal errors, heteroskedasticity, autocorrelation, censored, multinomial and simultaneous response data, as well as model selection and non-stationary and dynamic control problems Maximum Entropy Econometrics will be of interest to econometricians trying to devise procedures for recovering information from partial or incomplete data, as well as quantitative economists in finance and business, statisticians, and students and applied researchers in econometrics, engineering and the physical sciences.
Amos Golan is a professor of economics and directs the Info-Metrics Institute at American University. He is also an External Professor at the Santa Fe Institute and a Senior Associate at Pembroke College, Oxford. His research is primarily in the interdisciplinary field of info-metrics - the science and practice of information processing, modeling, inference, and problem solving with insufficient information. He has published in economics, econometrics, statistics, mathematics, physics and philosophy journals. His books include Maximum Entropy Econometrics: Robust Estimation with Limited Data (coauthored with Judge and Miller) and Information and Entropy Econometrics - A Review and Synthesis.
Inhaltsangabe
The Classical Maximum Entropy Formalism: A Review.
PURE INVERSE PROBLEMS.
Basic Maximum Entropy Principle: Formulation and Extensions.
Formulation and Solution of Pure Inverse Problems.
Generalized Pure Inverse Problems.
LINEAR INVERSE PROBLEMS WITH NOISE.
Generalized Maximum Entropy (GME) and Cross-Entropy (GCE)Formulations.
Finite Sample Extensions of GME-GCE.
GENERAL LINEAR MODEL APPLICATIONS OF GME-GCE.
GME-GCE Solutions to Ill-conditioned Problems.
General Linear Statistical Model with a Non-scalar IdentityCovariance Matrix Statistical Model Selection.
A SYSTEM OF ECONOMIC STATISTICAL RELATIONS.
Sets of Linear Statistical Models.
Simultaneous Equations Statistical Model.
LINEAR AND NON-LINEAR DYNAMIC SYSTEMS.
Estimation and Inference of Dynamic Linear Inverse Problems.
Linear and Non-linear Dynamic Systems with Control.
DISCRETE CHOICE-CENSORED PROBLEMS.
Recovering Information from Multinomial Response Data.
Recovering Information from Censored Response Data.
