摘要
This paper utilizes a change-point estimator based on <span>the </span><span style="font-style:italic;">φ</span><span>-</span><span>divergence. Since </span><span "=""><span>we seek a </span><span>near perfect</span><span> translation to reality, then locations of parameter change within a finite set of data have to be accounted for since the assumption of </span><span>stationary</span><span> model is too restrictive especially for long time series. The estimator is shown to be consistent through asymptotic theory and finally proven through simulations. The estimator is applied to the generalized Pareto distribution to estimate changes in the scale and shape parameters.</span></span>
This paper utilizes a change-point estimator based on <span>the </span><span style="font-style:italic;">φ</span><span>-</span><span>divergence. Since </span><span "=""><span>we seek a </span><span>near perfect</span><span> translation to reality, then locations of parameter change within a finite set of data have to be accounted for since the assumption of </span><span>stationary</span><span> model is too restrictive especially for long time series. The estimator is shown to be consistent through asymptotic theory and finally proven through simulations. The estimator is applied to the generalized Pareto distribution to estimate changes in the scale and shape parameters.</span></span>
作者
Mwelu Susan
Anthony G. Waititu
Peter N. Mwita
Charity Wamwea
Mwelu Susan;Anthony G. Waititu;Peter N. Mwita;Charity Wamwea(Pan-African University Institute of Basic Sciences, Technology and Innovation, Nairobi, Kenya;Department of Statistics and Actuarial Sciences, JKUAT, Nairobi, Kenya;Department of Mathematics, Statistics and Actuarial Sciences, Machakos University, Machakos, Kenya)
