Robust Non‐Nested Testing for Ordinary Least Squares Regression When Some of the Regressors are Lagged Dependent Variables

The problem of testing nonnested regression models that include lagged values of the dependent variable as regressors is discussed. It is argued that it is essential to test for error autocorrelation if ordinary least squares and the associated J and F tests are to be used. A heteroskedasticity-robust joint test against a combination of the artificial alternatives used for autocorrelation and nonnested hypothesis tests is proposed. Monte Carlo results indicate that implementing this joint test using a wild bootstrap method leads to a well-behaved procedure and gives better control of finite sample significance levels than asymptotic critical values.

[1]  Nicholas M. Kiefer,et al.  Simple Robust Testing of Regression Hypotheses , 2000 .

[2]  Li Qi,et al.  Bootstrapping J-type tests for non-nested regression models , 1995 .

[3]  J. MacKinnon,et al.  Heteroskedasticity-Robust Tests in Regression Directions , 1985 .

[4]  Bruce E. Hansen,et al.  Discussion of 'Data mining reconsidered' , 1999 .

[5]  Jean-Francois Richard,et al.  The Encompassing Principle and Its Application to Testing Non-nested Hypotheses , 1986 .

[6]  L. G. Godfrey,et al.  The wild bootstrap and heteroskedasticity-robust tests for serial correlation in dynamic regression models , 2005, Comput. Stat. Data Anal..

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

[8]  Francisco Cribari Neto.  Asymptotic inference under heteroskedasticity of unknown form , 2004 .

[9]  Emmanuel Flachaire,et al.  The wild bootstrap, tamed at last , 2001 .

[10]  Michael McAleer,et al.  The significance of testing empirical non-nested models , 1995 .

[11]  R. Serlin,et al.  Testing for robustness in Monte Carlo studies. , 2000, Psychological methods.

[12]  J. S. Long,et al.  Using Heteroscedasticity Consistent Standard Errors in the Linear Regression Model , 2000 .

[13]  James Davidson,et al.  Implementing the wild bootstrap using a two-point distribution☆ , 2007 .

[14]  Nicholas M. Kiefer,et al.  HETEROSKEDASTICITY-AUTOCORRELATION ROBUST TESTING USING BANDWIDTH EQUAL TO SAMPLE SIZE , 2002, Econometric Theory.

[15]  J. Darroch,et al.  On Testing More than One Hypothesis , 1963 .

[16]  L. G. Godfrey,et al.  Alternative approaches to implementing Lagrange multiplier tests for serial correlation in dynamic regression models , 2007, Comput. Stat. Data Anal..

[17]  J. MacKinnon,et al.  Several Tests for Model Specication in the Pres-ence of Alternative Hypotheses , 1981 .

[18]  Leslie Godfrey,et al.  Tests of non-nested regression models: Small sample adjustments and Monte Carlo evidence , 1983 .

[19]  Joel L. Horowitz,et al.  Empirically relevant critical values for hypothesis tests: A bootstrap approach , 2000 .

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

[21]  H. White,et al.  Some heteroskedasticity-consistent covariance matrix estimators with improved finite sample properties☆ , 1985 .

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

[23]  A. R. Tremayne,et al.  Checks of model adequacy for univariate time series models and their application to econometric relationships , 1998 .

[24]  James G. MacKinnon,et al.  Heteroskedastcity-robust tests in regressions directions , 1985 .

[25]  H. White.  A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity , 1980 .

[26]  J. Wooldridge.  Introduction to Econometrics , 2013 .

[27]  Nicholas M. Kiefer,et al.  HETEROSKEDASTICITY-AUTOCORRELATION ROBUST STANDARD ERRORS USING THE BARTLETT KERNEL WITHOUT TRUNCATION , 2002 .

[28]  David A. Hsieh,et al.  A heteroscedasticity-consistent covariance matrix estimator for time series regressions , 1983 .

[29]  Leslie Godfrey,et al.  Tests of non-nested regression models some results on small sample behaviour and the bootstrap , 1998 .

[30]  A. Bose.  Edgeworth correction by bootstrap in autoregressions , 1988 .

[31]  Grayham E. Mizon,et al.  A simple message for autocorrelation correctors: Don't , 1995 .

[32]  Emmanuel Flachaire,et al.  Bootstrapping heteroskedastic regression models: wild bootstrap vs. pairs bootstrap , 2005, Comput. Stat. Data Anal..

[33]  Thanasis Stengos,et al.  Semiparametric Specification Testing of Non-nested Econometric Models , 1994 .

[34]  Nicholas M. Kiefer,et al.  A NEW ASYMPTOTIC THEORY FOR HETEROSKEDASTICITY-AUTOCORRELATION ROBUST TESTS , 2005, Econometric Theory.

[35]  James G. MacKinnon,et al.  TESTS FOR MODEL SPECIFICATION IN THE PRESENCE OF ALTERNATIVE HYPOTHESES Some Further Results , 1983 .

[36]  Leslie Godfrey,et al.  Controlling the finite sample significance levels of heteroskedasticity-robust tests of several linear restrictions on regression coefficients , 2004 .

[37]  L. Godfrey.  TESTING AGAINST GENERAL AUTOREGRESSIVE AND MOVING AVERAGE ERROR MODELS WHEN THE REGRESSORS INCLUDE LAGGED DEPENDENT VARIABLES , 1978 .

[38]  D. Andrews.  Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation , 1991 .

[39]  M. Hashem Pesaran,et al.  Non-nested Hypothesis Testing: An Overview , 1999 .

[40]  T. Breusch.  TESTING FOR AUTOCORRELATION IN DYNAMIC LINEAR MODELS , 1978 .

[41]  N. Kiefer,et al.  Robust Nonnested Testing and the Demand for Money , 2008 .

[42]  J. MacKinnon,et al.  The interpretation of test statistics , 1985 .