A comparison of several exact and approximate tests for structural shift under heteroscedasticity

Abstract It is known that the Chow test for equality of regression coefficients is not robust to heteroscedasticity. Much recent attention has been paid to this problem by way of new tests. The bulk of the new procedures are justified only asymptotically and little has been done to examine finite sample properties. Even less attention has been focused on comparisons of tests under either the null or alternative hypotheses. In this paper we redress these shortcomings and also contribute by suggesting additional tests. The results indicate that there are tests available with good sampling characteristics, some of which are computationally easy to carry out from both the standpoint of the researcher's time and computation time.

[1]  Yuzo Honda,et al.  MODIFIED WALD TESTS IN TESTS OF EQUALITY BETWEEN SETS OF COEFFICIENTS IN TWO LINEAR REGRESSIONS UNDER HETEROSCEDASTICITY , 1986 .

[2]  Robin C. Sickles,et al.  Some Further Evidence on the Use of the Chow Test under Heteroskedasticity , 1977 .

[3]  H. Tsurumi,et al.  Some tests for the constancy of regressions under heteroscedasticity , 1985 .

[4]  G. Chow.  Tests of equality between sets of coefficients in two linear regressions (econometrics voi 28 , 1960 .

[5]  A. Zellner.  An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation Bias , 1962 .

[6]  H. Tsurumi.  ON JAYATISSA'S TEST OF CONSTANCY OF REGRESSIONS UNDER HETEROSCEDASTICITY AND SOME ALTERNATIVE TEST PROCEDURES , 1984 .

[7]  Yuzo Honda.  ON TESTS OF EQUALITY BETWEEN SETS OF COEFFICIENTS IN TWO LINEAR REGRESSIONS WHEN DISTURBANCE VARIANCES ARE UNEQUAL , 1982 .

[8]  Satterthwaite Fe.  An approximate distribution of estimates of variance components. , 1946 .

[9]  Toshihisa Toyoda,et al.  Testing equality between sets of coefficients after a preliminary test for equality of disturbance variances in two linear regressions , 1986 .

[10]  Michael D. Conerly,et al.  An Approximate Test for Comparing Heteroscedastic Regression Models , 1988 .

[11]  Toshihisa Toyoda,et al.  Use of the Chow Test under Heteroscedasticity , 1974 .

[12]  E. R. Mansfield,et al.  An approximate test for comparing independent regression models with unequal error variances , 1989 .

[13]  S. Goldfeld,et al.  Some Tests for Homoscedasticity , 1965 .

[14]  Masahito Kobayashi,et al.  A Bounds Test for Equality Between Sets of Coefficients in Two Linear Regression Models Under Heteroscedasticity , 1986, Econometric Theory.

[15]  Toshihisa Toyoda,et al.  Small Sample Properties of Tests of Equality between Sets of Coefficients in Two Linear Regressions under Heteroscedasticity , 1985 .

[16]  P. Phillips.  The Exact Distribution of the Wald Statistic , 1986 .

[17]  Masahito Kobayashi,et al.  A bounds test of equality between sets of coefficients in two linear regressions when disturbance variances are unequal , 1986 .

[18]  Thomas J Rothernberg.  HYPOTHESIS TESTING IN LINEAR MODELS WHEN THE ERROR COVARIANCE MATRIX IS NONSCALAR , 1984 .

[19]  Mukhtar M. Ali,et al.  Tests for Equality between Sets of Coefficients in Two Linear Regressions under Heteroscedasticity , 1985 .

[20]  Peter A. Watt,et al.  TESTS OF EQUALITY BETWEEN SETS OF COEFFICIENTS IN TWO LINEAR REGRESSIONS WHEN DISTURBANCE VARIANCES ARE UNEQUAL: SOME SMALL SAMPLE PROPERTIES* , 1979 .

[21]  S. Dalal,et al.  A conservative test and confidence region for comparing heteroscedastic regressions , 1988 .

[22]  J. MacKinnon,et al.  Efficient estimation of tail-area probabilities in sampling experiments , 1981 .

[23]  Samaradasa Weerahandi,et al.  TESTING REGRESSION EQUALITY WITH UNEQUAL VARIANCES , 1987 .

[24]  W. Jayatissa.  Tests of Equality between Sets of Coefficients in Two Linear Regressions when Disturbance Variances Are Unequal , 1977 .

[25]  Michael D. Perlman,et al.  Power of the Noncentral F-Test: Effect of Additional Variates on Hotelling's T2-Test , 1974 .