Time Series Models with a Specified Symmetric Non-Normal Marginal Distribution.
NAVAL POSTGRADUATE SCHOOL MONTEREY CA
Pagination or Media Count:
Time series models with autoregressive, moving average and mixed autoregressive-moving averager correlation structure and with symmetric, heavy-tailed, non-normal marginal distributions, called letter-Laplace, are considered. First, a flexible mixed model NLARMAp,q with Laplace double exponential marginals is investigated. Second, a family of continuous random coefficient models with l-Laplace distributions are examined. The Laplace distribution is described along with a useful transformation. Thirdly, the NLAR1 and the BELAR1 processes are compared using higher order residual analyses based on the uncorrelated, but dependent linear residuals, R sub n. Finally, open problems, as well as possible extensions and applications of the analyses given in this thesis are discussed. Keywords Maximum Likelihood estimation Least squares method Residual analysis.
- Statistics and Probability