Properties of Optimal Forecasts under Asymmetric Loss and Nonlinearity

Abstract Evaluation of forecast optimality in economics and finance has almost exclusively been conducted under the assumption of mean squared error loss. Under this loss function optimal forecasts should be unbiased and forecast errors serially uncorrelated at the single period horizon with increasing variance as the forecast horizon grows. Using analytical results we show that standard properties of optimal forecasts can be invalid under asymmetric loss and nonlinear data generating processes and thus may be very misleading as a benchmark for an optimal forecast. We establish instead that a suitable transformation of the forecast error—known as the generalized forecast error—possesses an equivalent set of properties. The paper also provides empirical examples to illustrate the significance in practice of asymmetric loss and nonlinearities and discusses the effect of parameter estimation error on optimal forecasts.

[1]  F. Diebold,et al.  Further results on forecasting and model selection under asymmetric loss , 1996 .

[2]  C. Granger,et al.  Economic and Statistical Measures of Forecast Accuracy , 1999 .

[3]  T. Bollerslev,et al.  Generalized autoregressive conditional heteroskedasticity , 1986 .

[4]  Allan Timmermann,et al.  Estimation and Testing of Forecast Rationality under Flexible Loss , 2005 .

[5]  T. Nijman,et al.  Temporal Aggregation of GARCH Processes. , 1993 .

[6]  J. Neyman,et al.  Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability , 1963 .

[7]  W. Rudin.  Principles of mathematical analysis , 1964 .

[8]  S. Satchell,et al.  Cost of Capital and the Regulator’s Preferences: An Investigation into a New Method of Estimating Regulatory Beta , 2004 .

[9]  Jan R. Magnus,et al.  The Bias of Forecasts from a First-Order Autoregression , 1991, Econometric Theory.

[10]  BAYESIAN MODELING OF ECONOMIES AND DATA REQUIREMENTS , 2001, Macroeconomic Dynamics.

[11]  Jan R. Magnus,et al.  The exact multi-period mean-square forecast error for the first-order autoregressive model , 1988 .

[12]  A. Robert Nobay,et al.  Optimal Discretionary Monetary Policy in a Model of Asymmetric Central Bank Preferences , 2003 .

[13]  Spyros Skouras,et al.  Decisionmetrics: A Decision-Based Approach to Econometric Modelling , 2007 .

[14]  Peter Christoffersen,et al.  Série Scientifique Scientific Series the Importance of the Loss Function in Option Valuation the Importance of the Loss Function in Option Valuation , 2022 .

[15]  A. Zellner.  Bayesian Estimation and Prediction Using Asymmetric Loss Functions , 1986 .

[16]  S. Satchell,et al.  Cost of Capital and Regulator’s Preferences: Investigation into a new method of estimating regulatory bias , 2004 .

[17]  James D. Hamilton.  A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle , 1989 .

[18]  Allan Timmermann,et al.  Optimal Forecast Combinations Under General Loss Functions and Forecast Error Distributions , 2002 .

[19]  A. Zellner,et al.  Forecasting international growth rates using Bayesian shrinkage and other procedures , 1989 .

[20]  Francis X. Diebold,et al.  Elements of Forecasting , 1997 .

[21]  Anthony D. Hall,et al.  Using Bayesian Variable Selection Methods to Choose Style Factors in Global Stock Return Models , 2000 .

[22]  Kenneth J. White,et al.  Introduction to the theory and practice of econometrics--a computer handbook using SHAZAM and SAS : for use with Judge-Hill-Griffiths-Lütkepohl-Lee, Introduction to the theory and practice of econometrics, second edition , 1988 .

[23]  H. L. Le Roy,et al.  Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability; Vol. IV , 1969 .

[24]  Enrique Sentana.  Least Squares Predictions and Mean-Variance Analysis , 2005 .

[25]  Norman R. Swanson,et al.  A Consistent Test for Nonlinear Out of Sample Predictive Accuracy , 2002 .

[26]  F. Diebold,et al.  Optimal Prediction Under Asymmetric Loss , 1994, Econometric Theory.

[27]  Chang‐Jin Kim,et al.  State Space Models with Regime Switching , 1999 .

[28]  Stephen E. Fienberg,et al.  Studies in Bayesian econometrics and statistics : in honor of Leonard J. Savage , 1975 .

[29]  G. Judge,et al.  The Theory and Practice of Econometrics , 1981 .

[30]  Peter Schmidt,et al.  The Theory and Practice of Econometrics , 1985 .

[31]  P. Phillips.  BOOTSTRAPPING I(1) DATA BY PETER C. B. PHILLIPS COWLES FOUNDATION PAPER NO. 1310 COWLES FOUNDATION FOR RESEARCH IN ECONOMICS , 2010 .

[32]  E. L. Lehmann,et al.  Theory of point estimation , 1950 .

[33]  Jan R. Magnus,et al.  The exact multi-period mean-square forecast error for the first-order autoregressive model , 1988 .

[34]  C. Whiteman.  Bayesian Prediction Under Asymmetric Linear Loss: Forecasting State Tax Revenues in Iowa , 1996 .

[35]  Chang‐Jin Kim,et al.  State-Space Models with Regime-Switching: Classical and Gibbs Sampling Approaches with Applications , 1999 .

[36]  C. Granger,et al.  Forecasting Economic Time Series. , 1988 .

[37]  Roy Batchelor,et al.  Rationality testing under asymmetric loss , 1998 .

[38]  James D. Hamilton.  Time Series Analysis , 1994 .

[39]  C. Stein,et al.  Estimation with Quadratic Loss , 1992 .

[40]  Tamar Frankel.  [The theory and the practice...]. , 2001, Tijdschrift voor diergeneeskunde.

[41]  Andrew J. Patton,et al.  Testing Forecast Optimality Under Unknown Loss , 2007 .

[42]  John M. Olin.  Calculating posterior distributions and modal estimates in Markov mixture models , 1996 .

[43]  S. Chib,et al.  Bayes inference via Gibbs sampling of autoregressive time series subject to Markov mean and variance shifts , 1993 .

[44]  Andrew Chesher,et al.  Residual analysis in the grouped and censored normal linear model , 1987 .

[45]  Clive W. J. Granger,et al.  Outline of forecast theory using generalized cost functions , 1999 .

[46]  M. Rosenblatt.  Remarks on a Multivariate Transformation , 1952 .

[47]  F. Diebold,et al.  Forecast Evaluation and Combination , 1996 .