Adaptive Estimation of Autoregressive Models with Time-Varying Variances

Stable autoregressive models of known finite order are considered with martingale differences errors scaled by an unknown nonparametric time-varying function generating heterogeneity. An important special case involves structural change in the error variance, but in most practical cases the pattern of variance change over time is unknown and may involve shifts at unknown discrete points in time, continuous evolution or combinations of the two. This paper develops kernel-based estimators of the residual variances and associated adaptive least squares (ALS) estimators of the autoregressive coefficients. These are shown to be asymptotically efficient, having the same limit distribution as the infeasible generalized least squares (GLS). Comparisons of the efficient procedure and ordinary least squares (OLS) reveal that least squares can be extremely inefficient in some cases while nearly optimal in others. Simulations show that, when least squares work well, the adaptive estimators perform comparably well, whereas when least squares work poorly, major efficiency gains are achieved by the new estimators.

[1]  Guido M. Kuersteiner,et al.  Efficiency IV Estimation for Autoregressive Models with Conditional Heterogeneity , 2000 .

[2]  Margaret Mary McConnell,et al.  Output Fluctuations in the United States: What Has Changed Since the Early 1980s? , 1998 .

[3]  Gabriel Perez-Quiros,et al.  Output Fluctuations in the United States: What Has Changed Since the Early 1980s? , 1998 .

[4]  Asymptotic and Bootstrap Inference for AR( Infinite ) Processes with Conditional Heteroskedasticity , 2003 .

[5]  Lutz Kilian,et al.  Bootstrapping Autoregressions with Conditional Heteroskedasticity of Unknown Form , 2002, SSRN Electronic Journal.

[6]  J. Davidson.  Stochastic Limit Theory: An Introduction for Econometricians , 1994 .

[7]  Why does the GARCH(1,1) model fail to provide sensible longer- horizon volatility forecasts? , 2005 .

[8]  R. C. Merton,et al.  On Estimating the Expected Return on the Market: An Exploratory Investigation , 1980 .

[9]  Nonstationary nonlinear heteroskedasticity in regression , 2005 .

[10]  E. Mammen.  Bootstrap and Wild Bootstrap for High Dimensional Linear Models , 1993 .

[11]  Richard A. Davis,et al.  Time Series: Theory and Methods , 2013 .

[12]  Peter C. B. Phillips,et al.  An Asymptotic Theory of Bayesian Inference for Time Series , 1996 .

[13]  P. Perron,et al.  Estimating and testing linear models with multiple structural changes , 1995 .

[14]  B. Abraham,et al.  Inferences about the parameters of a time series model with changing variance , 1984 .

[15]  P. Phillips,et al.  Testing the covariance stationarity of heavy-tailed time series: An overview of the theory with applications to several financial datasets , 1994 .

[16]  Testing stationarity under a permanent variance shift , 2004 .

[17]  Giuseppe Cavaliere,et al.  Unit Root Tests under Time-Varying Variances , 2005 .

[18]  H. Müller.  CHANGE-POINTS IN NONPARAMETRIC REGRESSION ANALYSIS' , 1992 .

[19]  Clive W. J. Granger,et al.  Non-stationarities in stock returns , 2004 .

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

[21]  J. Durbin.  Estimation of Parameters in Time‐Series Regression Models , 1960 .

[22]  D. Andrews.  Laws of Large Numbers for Dependent Non-Identically Distributed Random Variables , 1988, Econometric Theory.

[23]  Yuichi Kitamura,et al.  Empirical Likelihood Based Inference in Conditional Moment Restriction Models , 2004 .

[24]  Dean W. Wichern,et al.  Changes of Variance in First-Order Autoregressive Time Series Models -With an Application , 1976 .

[25]  N. Shephard.  Stochastic Volatility: Selected Readings , 2005 .

[26]  H. P. Boswijk.  Adaptive Testing for a Unit Root with Nonstationary Volatility , 2005 .

[27]  R. Miller,et al.  On the Stable Paretian Behavior of Stock-Market Prices , 1974 .

[28]  Daniel Peña,et al.  Covariance changes detection in multivariate time series , 2007 .

[29]  Francis X. Dieobold.  Modeling The persistence Of Conditional Variances: A Comment , 1986 .

[30]  R. Engle.  Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation , 1982 .

[31]  R. Officer.  The Variability of the Market Factor of the New York Stock Exchange. , 1973 .

[32]  H. White.  Asymptotic theory for econometricians , 1985 .

[33]  P. Robinson,et al.  Nonparametric Estimation of Time-Varying Parameters , 1989 .

[34]  R. Engle,et al.  The Spline GARCH Model for Unconditional Volatility and its Global Macroeconomic Causes , 2005 .

[35]  M. Watson.  Explaining the Increased Variability in Long-Term Interest Rates , 1999 .

[36]  Miguel A. Delgado,et al.  Nonparametric inference on structural breaks , 2000 .

[37]  W. Wong.  On the Consistency of Cross-Validation in Kernel Nonparametric Regression , 1983 .

[38]  M. C. Jones,et al.  Likelihood-Based Local Linear Estimation of the Conditional Variance Function , 2004 .

[39]  Peter C. B. Phillips,et al.  Econometric Model Determination , 1996 .

[40]  J. Bai,et al.  Least squares estimation of a shift in linear processes , 1994 .

[41]  K. French,et al.  Expected stock returns and volatility , 1987 .

[42]  Sangyeol Lee,et al.  The Cusum of Squares Test for Scale Changes in Infinite Order Moving Average Processes , 2001 .

[43]  A. M. Robert Taylor,et al.  Variance Shifts, Structural Breaks, and Stationarity Tests , 2003 .

[44]  P. Phillips,et al.  Testing Covariance Stationarity Under Moment Condition Failure with an Application to Common Stock Returns , 1990 .

[45]  R. Tsay.  Outliers, Level Shifts, and Variance Changes in Time Series , 1988 .

[46]  EFFICIENT IV ESTIMATION FOR AUTOREGRESSIVE MODELS WITH CONDITIONAL HETEROSKEDASTICITY , 2002, Econometric Theory.

[47]  Raymond J. Carroll,et al.  Adapting for Heteroscedasticity in Linear Models , 1982 .

[48]  Christian M. Hafner,et al.  Testing for Linear Autoregressive Dynamics under Heteroskedasticity , 2000 .

[49]  Peter C. B. Phillips,et al.  ASYMPTOTICS FOR NONLINEAR TRANSFORMATIONS OF INTEGRATED TIME SERIES , 1999, Econometric Theory.

[50]  Michael McAleer,et al.  A Survey of Recent Theoretical Results for Time Series Models with GARCH Errors , 2001 .

[51]  Regina Y. Liu.  Bootstrap Procedures under some Non-I.I.D. Models , 1988 .

[52]  A. M. Robert Taylor,et al.  Testing for unit roots in time series models with non-stationary volatility , 2007 .

[53]  Richard A. Davis,et al.  Time Series: Theory and Methods (2nd ed.). , 1992 .

[54]  C. Stărică,et al.  A simple non-stationary model for stock returns , 2002 .

[55]  Changbao Wu,et al.  Jackknife, Bootstrap and Other Resampling Methods in Regression Analysis , 1986 .

[56]  M. Kupperman.  Linear Statistical Inference and Its Applications 2nd Edition (C. Radhakrishna Rao) , 1975 .

[57]  P. Robinson.  Asymptotically efficient estimation in the presence of heteroskedasticity of unknown form , 1987 .

[58]  B. Hansen.  REGRESSION WITH NONSTATIONARY VOLATILITY , 1995 .

[59]  Keith D. C. Stoodley,et al.  Time Series Analysis: Theory and Practice I , 1982 .

[60]  P. Phillips,et al.  Inference in Autoregression under Heteroskedasticity , 2006 .

[61]  T. Mikosch,et al.  Nonstationarities in Financial Time Series, the Long-Range Dependence, and the IGARCH Effects , 2004, Review of Economics and Statistics.

[62]  Peter C. B. Phillips,et al.  Nonlinear Regressions with Integrated Time Series , 2001 .

[63]  Jianqing Fan,et al.  Efficient Estimation of Conditional Variance Functions in Stochastic Regression , 1998 .

[64]  Paul Newbold,et al.  Unit root tests with a break in innovation variance , 2002 .

[65]  A. Bose.  Bootstrap in moving average models , 1990 .

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

[67]  G. C. Tiao,et al.  Use of Cumulative Sums of Squares for Retrospective Detection of Changes of Variance , 1994 .

[68]  H. Müller,et al.  Local Polynomial Modeling and Its Applications , 1998 .

[69]  Shigeyuki Hamori,et al.  Testing for a unit root in the presence of a variance shift1 , 1997 .

[70]  Optimal Instrumental Variables Estimation For ARMA Models , 2001 .

[71]  C. Granger,et al.  Nonstationarities in Stock Returns , 2005, Review of Economics and Statistics.

[72]  D. Dijk,et al.  Testing for changes in volatility in heteroskedastic time series - a further examination , 2004 .

[73]  Dick van Dijk,et al.  Changes in variability of the business cycle in the G7 countries , 2002 .

[74]  Giuseppe Cavaliere,et al.  STATIONARITY TESTS UNDER TIME-VARYING SECOND MOMENTS , 2005, Econometric Theory.

[75]  Professors Engle,et al.  MODELING THE PERSISTENCE OF CONDITIONAL VARIANCES: A COMMENT , 1986 .

[76]  J. Durbin,et al.  Techniques for Testing the Constancy of Regression Relationships Over Time , 1975 .

[77]  C. R. Rao,et al.  Linear Statistical Inference and its Applications , 1968 .

[78]  Lutz Kilian,et al.  Asymptotic and Bootstrap Inference for AR(∞) Processes with Conditional Heteroskedasticity , 2007 .

[79]  Y. Yin,et al.  Detection of the number, locations and magnitudes of jumps , 1988 .

[80]  Jens-Peter Kreiss.  Asymptotical Properties of Residual Bootstrap for Autoregressions , 1997 .

[81]  C. Chu,et al.  Kernel-Type Estimators of Jump Points and Values of a Regression Function , 1993 .

[82]  Varying coefficient GARCH versus local constant volatility modeling: comparison of the predictive power , 2006 .

[83]  P. Robinson,et al.  Time-Varying Nonlinear Regression , 1991 .

[84]  Ching-Kang Ing,et al.  MULTISTEP PREDICTION IN AUTOREGRESSIVE PROCESSES , 2003, Econometric Theory.

[85]  V. P. Godambe.  An Optimum Property of Regular Maximum Likelihood Estimation , 1960 .

[86]  H. P. Boswijk,et al.  Testing for a Unit Root with Near-Integrated Volatility , 2001 .

[87]  EFFICIENT ESTIMATION OF NONSTATIONARY TIME SERIES REGRESSION , 1988 .

[88]  J. Franke,et al.  BOOTSTRAPPING STATIONARY AUTOREGRESSIVE MOVING-AVERAGE MODELS , 1992 .

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

[90]  C. Stărică.  Is GARCH(1,1) as good a model as the Nobel prize accolades would imply?1 , 2003 .