Empirical Modeling in Economics: Specification and EvaluationCambridge University Press, 30. 9. 1999 - Počet stran: 99 In these three essays, Professor Granger explains the process of constructing and evaluating an empirical model. Drawing on a wide range of cases and vignettes from economics, finance, politics and environment economics, as well as from art, literature, and the entertainment industry, Professor Granger combines rigor with intuition to provide a unique and entertaining insight into one of the most important subjects in modern economics. Chapter 1 deals with Specification. Chapter 2 considers Evaluation, and argues that insufficent evaluation is undertaken by economists, and that models should be evaluated in terms of the quality of their output. In Chapter 3, the question of how to evaluate forecasts is considered at several levels of increasing depth. |
Obsah
The specification of empirical models | 1 |
The evaluation of empirical models | 33 |
Comments on the evaluation of econometric models and of forecasts | 61 |
Další vydání - Zobrazit všechny
Empirical Modeling in Economics: Specification and Evaluation Clive W. J. Granger Náhled není k dispozici. - 1999 |
Běžně se vyskytující výrazy a sousloví
actual economy aggregate Alfred Marshall alternative models Amazon region applied approach appropriate Cambridge chapter conditional distribution conditional mean considered consumer consumption corresponding cost function c(e criterion crop land D.F. Hendry decision maker deforestation difficult discussed econometric models econometricians economic theory economists elasticity empirical model estimates Evaluation of forecasts example expected value explanatory variables fallow land fore forecast errors Granger Hall's theory Hendry implications in-sample information set John Maynard Keynes lags linear Lucas critique Marshall Lectures newly cleared land nomic occur optimum outcomes output parameters particularly pasture percent Permanent Income Hypothesis Pesaran piece point forecast policy models possible post-sample prob probability problem properties question rain forest region of Brazil regression relevant risk neutral S₁ second-order stochastic dominance simple specification spurious relationship Suppose techniques theory or model tion users utility function zero