Abstract
A bstract-We discuss the problem of testing for constant versus time varying regression coefficients. Our alternative hypothesis allows the coefficients to follow a stationary AR(1) process with unknown autoregressive parameter. Standard testing procedures are inappropriate since this parameter is identified only under the alternative. We propose a test statistic which is a function of a sequence of Score statistics, and depends only on the regressors and the OLS residuals. The distribution of the test statistic is discussed, power and size are investigated using Monte Carlo methods, and an empirical example investigating stability in the gold and silver markets is presented.
Keywords
RegressionStability (learning theory)EconometricsStatisticsRegression analysisMathematicsLinear regressionEconomicsComputer scienceMachine learning
Related Publications
Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
Traditional econometric models assume a constant one-period forecast variance. To generalize this implausible assumption, a new class of stochastic processes called autoregressi...
On the misuse of residuals in ecology: regression of residuals vs. multiple regression
1 Residuals from linear regressions are used frequently in statistical analysis, often for the purpose of controlling for unwanted effects in multivariable datasets. This paper ...
Least Median of Squares Regression
Abstract Classical least squares regression consists of minimizing the sum of the squared residuals. Many authors have produced more robust versions of this estimator by replaci...
Empirical Problems of the Hierarchical Likelihood Ratio Test for Model Selection
Advocates of maximum likelihood (ML) approaches to phylogenetics commonly cite as one of their primary advantages the use of objective statistical criteria for model selection. ...
Stationarity and Persistence in the GARCH(1,1) Model
This paper establishes necessary and sufficient conditions for the stationarity and ergodicity of the GARCH(l.l) process. As a special case, it is shown that the IGARCH(1,1) pro...
Publication Info
- Year
- 1985
- Type
- article
- Volume
- 67
- Issue
- 2
- Pages
- 341-341
- Citations
- 96
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
96
OpenAlex
Cite This
Mark W. Watson,
Robert F. Engle
(1985).
Testing for Regression Coefficient Stability with a Stationary AR(1) Alternative.
The Review of Economics and Statistics
, 67
(2)
, 341-341.
https://doi.org/10.2307/1924737
Identifiers
- DOI
- 10.2307/1924737