Wavelet-Based Bayesian Analysis of Generalized Long-Memory Process
Alex Gonzaga, Akira Kawanaka · 2006
In this paper we propose a Bayesian approach to estimating the parameters and predicting future values of a generalized long-memory process utilizing the approximate likelihood function of discrete wavelet packet coefficients. This approximation does not depend on the length of the signal, but the length of the wavelet filter, which is under the control of the analyst. We illustrate our approach by an example applying simulated data.