|
| << vorige | volgende >> |
Tijdschrift beschrijving
Alle jaargangen van het bijbehorende tijdschrift
Alle afleveringen van het bijbehorende jaargang
Alle artikelen van de bijbehorende aflevering
Details van artikel 3 van 20 gevonden artikelen
| Titel: |
A corrected likelihood ratio statistic for the multivariate regression model with antedependent errors |
| Auteur: |
Albert, Jeffrey M. |
| Verschenen in: |
Communications in statistics |
| Paginering: | Jaargang 21 (1992) nr. 7 pagina's 1823-1843 |
| Jaar: |
1992 |
| Inhoud: |
This paper addresses the problem of inference for the antedependence model (Gabriel, 1961, 1962). Antedependence can be formulated as an autoregressive process of general order which is non-stationary in time. Its primary application is in the analysis of repeated measurements data, that is, data consisting of independent replicates of relatively short time series. Our focus is on testing a general linear hypothesis in the context of a multivariate regression model with multivariate normal antedependent errors. Although the relevant likelihood ratio statistic was first presented by Gabriel (1961), the distribution of this statistics has not yet been derived. We present this derivation and show how this result leads to a simple correction factor to improve the x2 approximation of the likelihood ratio statistic. |
| Uitgever: |
Taylor & Francis |
| Bronbestand: |
Elektronische Wetenschappelijke Tijdschriften |
Details van artikel 3 van 20 gevonden artikelen
| << vorige | volgende >> |