A method for fitting stable autoregressive models using the autocovariation function

✍ Scribed by Colin M Gallagher

• Publisher: Elsevier Science
• Year: 2001
• Tongue: English
• Weight: 189 KB
• Volume: 53
• Category: Article
• ISSN: 0167-7152
• DOI: 10.1016/s0167-7152(01)00041-4

✦ Synopsis

We use the sample covariations to estimate the parameters in a univariate symmetric stable autoregressive process. Unlike the sample correlation, the sample covariation can be used to estimate the tail decay parameter of the process. The ÿtted model will be consistent with the dependence as measured by the covariation. The limit distribution of the sample covariation can be used to derive conÿdence intervals for the autoregressive parameter in a ÿrst order process. Simulations show that conÿdence intervals coming from the covariation have better coverage probabilities than those coming from the sample correlations. The method is demonstrated on a time series of sea surface temperatures.