Pre-checking for non-monotonicity of the Wald statistic
- 1 September 2005
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 75 (9) , 751-759
Abstract
The non-monotonic behaviour of the Wald test in some finite-sample applications leads to low power when the null hypothesis needs rejection most. This article proposes a simple check for discerning if the Wald statistic for testing significance of regression coefficients is non-monotonic in the neighbourhood of the parameter space from which the sample data are drawn. Monte Carlo simulations show that this method works rather well for detecting situations where the Wald test can be safely applied. An example is provided to illustrate the use of this check.Keywords
This publication has 9 references indexed in Scilit:
- Modified Wald test for regression disturbancesEconomics Letters, 1997
- The Danger of Extrapolating Asymptotic Local PowerEconometrica, 1990
- Understanding Wald's Test for Exponential FamiliesThe American Statistician, 1987
- Understanding Wald's Test for Exponential FamiliesThe American Statistician, 1987
- On the Use of Wald's Test in Exponential FamiliesInternational Statistical Review, 1985
- Estimates of the Deterrent Effect of Capital Punishment: The Importance of the Researcher's Prior BeliefsJournal of Political Economy, 1985
- Corrigenda: Wald's Test As Applied to Hypotheses in Logit AnalysisJournal of the American Statistical Association, 1980
- Wald's Test as Applied to Hypotheses in Logit AnalysisJournal of the American Statistical Association, 1977
- Wald's Test as Applied to Hypotheses in Logit AnalysisJournal of the American Statistical Association, 1977