Abstract
We produce Monte Carlo evidence on the size and power of the RESET, a heteroscedasticity test, and a test for autocorrelation applied to realistic distributed-lag models. We find that the autocorrelation test has the correct size and high power to detect not only autocorrelation (given a correct model), but also the erroneous omission of several lags of an explanatory variable, whereas the RESET and heteroscedasticity tests are oversized in the presence of positive disturbance autocorrelation, especially when the regressors are also positively autocorrelated, and have no power to detect such misspecification errors. In large samples, size distortion may be avoided by using autocorrelation-robust methods.
Similar content being viewed by others
Notes
We use the LINUX version of RATS v. 5.01 to carry out the simulations. Random numbers were generated using the function %RAN(x) and the starting seed 317811. Note also that before we started drawing the values of ɛ 1, ɛ 2, and w, we let the process run for 500 “periods.”
All the tables not reported here are available upon request.
In the first version of the paper, we only considered the omission of four lags of X 1t . We thank an anonymous referee of this journal for suggesting a generalization, so as to avoid conclusions that depend on a specific model.
References
Godfrey LG (1988) Misspecification tests in econometrics: the Lagrange multiplier principle and other approaches. Cambridge University Press, New York
Godfrey LG, McAleer M, McKenzie CR (1988) Variable addition and Lagrange multiplier tests for linear and logarithmic regression models. Rev Econ Stat 70:492–503
Godfrey LG, Orme CD (1994) The sensitivity of some general checks to omitted variables in the linear model. Int Econ Rev 35:489–506
Johnston J (1972) Econometric methods. 2nd Ed. McGraw-Hill, Tokyo
Kiviet JF (1986) On the rigour of some misspecification tests for modelling dynamic relationships. Rev of Econ Stud 53:241–261
Krämer W, Kiviet J, Breitung J (1990) The null distribution of the F-test in the linear regression model with autocorrelated disturbances. Statistica 50:503–509
Krämer W, Sonnberger H, Maurer J, Havlik P (1985) Diagnostic checking in practice. Rev Econ Stat 67:118–123
Leung SF, Yu S (2001) The sensitivity of the RESET tests to disturbance autocorrelation in regression residuals. Empir Econ 26:721–726
Pagan AR, Hall AD (1983) Diagnostic tests as residual analysis. Econom Rev 2:159–218
Porter RD, Kashyap AK (1984) Autocorrelation and the sensitivity of RESET. Econ Lett 14:229–233
Ramsey JB (1969) Tests for specification errors in classical linear least-squares regression analysis. J R Stat Soc, B 31:350–371
Thursby JG (1979) Alternative specification error tests: A comparative study. J Am Stat Assoc 74:222–225
Thursby JG (1989) A comparison of several specification error tests for a general alternative. Int Econ Rev 30:217–230
Thursby JG, Schmidt P (1977) Some properties of tests for specification error in a linear regression model. J Am Stat Assoc 72:635–641
Wooldridge JM (2003) Introductory econometrics: a modern approach. 2nd Ed., South-Western, Thomson
Author information
Authors and Affiliations
Corresponding author
Additional information
We gratefully acknowledge the constructive comments of an anonymous referee of this journal, which improved significantly the paper. We also thank our colleagues S. Symeonides and E. Zacharias for their comments. The usual disclaimer applies.
Rights and permissions
About this article
Cite this article
Hatzinikolaou, D., Stavrakoudis, A. Empirical size and power of some diagnostic tests applied to a distributed lag model. Empirical Economics 31, 631–643 (2006). https://doi.org/10.1007/s00181-005-0039-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00181-005-0039-0