Higher order asymptotic theory for time series analysis (Q1189362)

The book has the following chapters: 1. A survey of the first-order asymptotic theory for time series analysis. 2. Higher order asymptotic theory for Gaussian ARMA processes. 3. Validity of Edgeworth expansions in time series analysis. 4. Higher order asymptotic sufficiency, asymptotic ancillarity in time series analysis. 5. Higher order investigations for testing theory in time series analysis. 6. Higher order asymptotic theory for multivariate time series. 7. Some practical examples. The book arose from a series of papers which were published by the author in different journals during the last 10 years. The results are based on Edgeworth expansions for statistics derived from a random process, and Berry-Esseen theorems for quadratic forms. The author extends the concept of higher order asymptotic sufficiency to Gaussian ARMA models. A class of tests about a parameter of the ARMA process is proposed. The class contains the likelihood ratio, Wald, modified Wald and Rao tests as special cases. The investigation of multivariate time series is based on the asymptotic expansion for the distribution of functions of the eigenvalues of the sample covariance matrix. The higher order asymptotic theory is verified by some numerical studies in the last chapter.