A0792
Title: A skew transition distribution modeling for higher-order circular Markov processes
Authors: Hiroaki Ogata - Tokyo Metropolitan University (Japan) [presenting]
Takayuki Shiohama - Nanzan University (Japan)
Abstract: The purpose is to propose an extension of the higher-order Markov processes on the circle where an underlying binding density has a skewing structure. The structures for circular autocorrelation functions (CACF), circular partial autocorrelation functions (CPACF), and the spectral density function of the process are investigated. The maximum likelihood estimation for model parameters is considered, and its finite sample performances are investigated by numerical simulations. As a real data analysis, time series of wind directions is used for practical purposes.