Applications of the Generalized Likelihood to Time Series Analysis
Abstract
Model-critical procedures provide a means to scrutinize as assumed parametric statistical model by varying the way the data are processed for repeated fits to the model. The criticism of the data is accomplished using the generalized likelihood function for the assumed probability density of the data. The degree of criticism is controlled by a user specified constant, c. The model-critical parameter estimates are obtained by maximization of the generalized likelihood function. When c=O, no criticism is performed and maximum likelihood estimates are obtained. Model-critical estimation procedures are presented for univariate and multivariate autoregressive-moving average processes. The procedures use a Kalman filter in evaluating the generalized likelihood function. A model selection criterion, based on the generalized likelihood, is also presented. A statistical test of fit for multivariate Gaussianity is presented; the test compares the model-critical and maximum likelihood estimates of the covariance matrix. Keywords: Statistical data, Tables data.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 28, 1988
- Accession Number
- ADA199101
Entities
People
- Gerald R. Swope
Organizations
- Naval Underwater Systems Center