Abstract
The asymptotic distribution of residual autocorrelations and a score statistic are derived for checking model adequacy for some Markov regression models for time series. These models are natural extensions of the popular generalized linear models to time series data. The score statistic appears to be somewhat easier to use in practice. Some empirical results are also obtained for the score statistic in some simple cases. A real example is given. Copyright © 1991 Oxford University Press on behalf of Biometrika Trust.
Original language | English |
---|---|
Pages (from-to) | 83-89 |
Journal | Biometrika |
Volume | 78 |
Issue number | 1 |
DOIs | |
Publication status | Published - Mar 1991 |
Citation
Li, W. K. (1991). Testing model adequacy for some Markov regression models for time series. Biometrika, 78(1), 83-89. doi: 10.1093/biomet/78.1.83Keywords
- Asymptotic distribution
- Generalized linear model
- Markov regression model
- Residual autocorrelation
- Score statistic