Premium
CONFIDENCE REGIONS FOR PARAMETERS IN THE AR(1) MODEL
Author(s) -
Hamilton David,
Wu Ka Ho
Publication year - 1995
Publication title -
journal of time series analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.576
H-Index - 54
eISSN - 1467-9892
pISSN - 0143-9782
DOI - 10.1111/j.1467-9892.1995.tb00233.x
Subject(s) - autoregressive model , mathematics , series (stratigraphy) , statistics , confidence interval , econometrics , confidence region , sample (material) , paleontology , biology , chemistry , chromatography
. The construction of approximate joint and marginal confidence regions for parameters in the first‐order autoregressive time series model is discussed. These regions are based on the large sample distributions of the likelihood ratio (and approximations to it), of the maximum likelihood estimates and of the score statistics. All these approaches are illustrated using a well‐known example from Box and Jenkins ( Time Series Analysis:Forecasting and Control , revised edn. San Francisco:Holden Day, 1976) and some simulated series. In addition, a simulation study is provided for comparing the coverage properties of the various procedures.