Better the Devil You Know: Improved Forecasts from Imperfect Models

Many important economic decisions are based on a parametric forecasting model that is known to be good but imperfect. We propose methods to improve out-of-sample forecasts from a mis-specified model by estimating its parameters using a form of local M estimation (thereby nesting local OLS and local MLE), drawing on information from a state variable that is correlated with the misspecification of the model. We theoretically consider the forecast environments in which our approach is likely to o¤er improvements over standard methods, and we find significant fore- cast improvements from applying the proposed method across distinct empirical analyses including volatility forecasting, risk management, and yield curve forecasting.

[1]  A. Timmermann,et al.  Conditional rotation between forecasting models , 2021, Journal of Econometrics.

[2]  Peter Reinhard Hansen,et al.  How should parameter estimation be tailored to the objective? , 2021, Journal of Econometrics.

[3]  A. Inoue,et al.  Local-Linear Estimation of Time-Varying-Parameter GARCH Models and Associated Risk Measures , 2021 .

[4]  Andrew J. Patton.  Comparing Possibly Misspecified Forecasts , 2020, Journal of Business & Economic Statistics.

[5]  Zhipeng Liao,et al.  Conditional Superior Predictive Ability , 2020, The Review of Economic Studies.

[6]  George Kapetanios,et al.  A similarity‐based approach for macroeconomic forecasting , 2019, Journal of the Royal Statistical Society: Series A (Statistics in Society).

[7]  James W. Taylor.  Forecasting Value at Risk and Expected Shortfall Using a Semiparametric Approach Based on the Asymmetric Laplace Distribution , 2019 .

[8]  Johanna F. Ziegel,et al.  Elicitability and backtesting: Perspectives for banking regulation , 2016, 1608.05498.

[9]  Zhaoyong Zhang,et al.  Weighted maximum likelihood for dynamic factor analysis and forecasting with mixed frequency data , 2016 .

[10]  Butt Man-Kit.  Forecasting Stock Returns under Economic Constraints , 2015 .

[11]  Johanna F. Ziegel,et al.  Higher order elicitability and Osband’s principle , 2015, 1503.08123.

[12]  R. Giacomini,et al.  Theory-coherent forecasting , 2014 .

[13]  Dennis Kristensen,et al.  Adding and subtracting Black-Scholes: A new approach to approximating derivative prices in continuous-time models , 2011 .

[14]  Andreas Pick,et al.  Optimal Forecasts in the Presence of Structural Breaks , 2011 .

[15]  Barbara Rossi,et al.  Forecast comparisons in unstable environments , 2010 .

[16]  Peter Reinhard Hansen,et al.  The Model Confidence Set , 2010 .

[17]  T. Gneiting.  Making and Evaluating Point Forecasts , 2009, 0912.0902.

[18]  Jianqing Fan,et al.  Local quasi-likelihood with a parametric guide. , 2009, Annals of statistics.

[19]  S. Manganelli.  Forecasting With Judgment , 2009 .

[20]  Fulvio Corsi,et al.  A Simple Approximate Long-Memory Model of Realized Volatility , 2008 .

[21]  Brendan K. Beare.  COPULAS AND TEMPORAL DEPENDENCE , 2008 .

[22]  Peter F. Christoffersen,et al.  Option Valuation with Long-Run and Short-Run Volatility Components , 2008 .

[23]  Piotr Cofta,et al.  The Model of Confidence , 2007 .

[24]  Jon Faust,et al.  Comparing Greenbook and Reduced Form Forecasts Using a Large Realtime Dataset , 2007 .

[25]  Allan Timmermann,et al.  Forecast Combination With Entry and Exit of Experts , 2006 .

[26]  Xiaohong Chen,et al.  Estimation of Copula-Based Semiparametric Time Series Models , 2006 .

[27]  P. Hansen,et al.  A Forecast Comparison of Volatility Models: Does Anything Beat a Garch(1,1)? , 2004 .

[28]  F. Diebold,et al.  Roughing It Up: Including Jump Components in the Measurement, Modeling, and Forecasting of Return Volatility , 2005, The Review of Economics and Statistics.

[29]  Min Wei,et al.  Do Macro Variables, Asset Markets or Surveys Forecast Inflation Better? , 2005 .

[30]  P. Hansen,et al.  A Forecast Comparison of Volatility Models: Does Anything Beat a Garch(1,1)? , 2004 .

[31]  Halbert White,et al.  Tests of Conditional Predictive Ability , 2003 .

[32]  Feifang Hu,et al.  The weighted likelihood , 2002 .

[33]  Wolfgang Karl Härdle,et al.  Nonparametric Vector Autoregression , 1998 .

[34]  Wolfgang Karl Härdle,et al.  Local polynomial estimators of the volatility function in nonparametric autoregression , 1997 .

[35]  Feifang Hu,et al.  The asymptotic properties of the maximum‐relevance weighted likelihood estimators , 1997 .

[36]  Andrew A. Weiss,et al.  Estimating Time Series Models Using the Relevant Cost Function , 1996 .

[37]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[38]  R. Engle,et al.  A Permanent and Transitory Component Model of Stock Return Volatility , 1993 .

[39]  W. Cleveland,et al.  Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting , 1988 .

[40]  R. Tibshirani,et al.  Local Likelihood Estimation , 1987 .

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

[42]  Clive W. J. Granger,et al.  Prediction with a generalized cost of error function , 1969 .

[43]  J. Muth.  Optimal Properties of Exponentially Weighted Forecasts , 1960 .

[44]  Andrew J. Patton Johanna F. Ziegel Rui Chen.  Supplemental Appendix to : Dynamic Semiparametric Models for Expected Shortfall ( and Value-at-Risk ) , 2017 .

[45]  Lu Jin,et al.  Rolling window selection for out-of-sample forecasting with time-varying parameters , 2017 .

[46]  Andrew J. Patton,et al.  Comparing Possibly Misspeci ed Forecasts , 2014 .

[47]  F. Diebold,et al.  The Rodney L . White Center for Financial Research Forecasting the Term Structure of Government Bond Yields , 2003 .

[48]  Jianqing Fan,et al.  Local maximum likelihood estimation and inference , 1998 .

[49]  A. Siegel,et al.  Parsimonious modeling of yield curves , 1987 .

[50]  W. Newey,et al.  Large sample estimation and hypothesis testing , 1986 .