Title:Uncertainty,information, and disagreement ofeconomic forecasters
Lecturer:Professor Ehsan S.Soofi,Lubar School of Business, University of Wisconsin-Milwaukee
Time:2017年11月27日10:00-12:00
Location:A618
Invited by Weiguo Fang
Abstract:
An information framework is proposed for studying uncertainty anddisagreement of economic forecasters. This framework builds upon the mixture model of combining density forecasts through a systematic application of the information theory. The framework encompasses the measuresused in the literature and leads to their generalizations. The focal measureis the Jensen–Shannon divergence of the mixture which admits Kullback–Leibler and mutual information representations. Illustrations include exploringthe dynamics of the individual and aggregate uncertainty about the USinflation rate using the survey of professional forecasters (SPF). We showthat the normalized entropy index corrects some of the distortions causedby changes of the design of the SPF over time. Bayesian hierarchical modelsare used to examine the association of the inflation uncertainty with theanticipated inflation and the dispersion of point forecasts.Implementationof the information framework based on the variance and Dirichlet modelfor capturing uncertainty about the probability distribution of the economicvariable are briefly discussed.
About the lecturer:
Ehsan S. Soofi is a University of Wisconsin-Milwaukee Distinguished Professor at the Lubar School of Business. He is an elected member of International Statistical Institute (ISI), and a Fellow of the American Statistical Association.Professor Soofi’s research interests are in information-theoretic and Bayesian pproaches to distribution theory and statistics, and their applications in reliability, economics and management sciences. He has published widely in the leading journals of statistics, econometrics,applied probability and operations research, engineering,management,information systems, and marketing.He is an Associate Editor of Econometric Reviews and served as the Guest Editor ofEconometric Reviews,Special Issue on Bayesian Inference and Information: In Memory of ArnoldZellner and as an Associate Editor of theJournal of the AmericanStatistical Association (1990-2005). He served as the Chair of the Leonard J. Savage Thesis Award Committee (1992-2002), as the Chair of the Industrial Statistics Section of the International Society for Bayesian Analysis (ISBA), as the Secretary (1997-1999), Vice President (1999-2001) and the Chair of Publication Committee (2001-2005) of International Association for Statistical Computing, and on the ISI Publication Committee (2003-05). Dr.Soofireceived his Ph.D. in Applied Statistics from the University of California,Riverside, MA in statistics from the University of California,Berkeley, and BA in mathematics from UCLA.
