The modified Yule-Walker method for -stable time series models

详细信息    查看全文

• 作者：Piotr Kruczek^a ; ^{piotr.kruczek@pwr.edu.pl} ; Agnieszka Wyłomańska^b ; ^{awylomanska@cuprum.wroc.pl} ; Marek Teuerle^a ; ^{marek.teuerle@pwr.edu.pl} ; Janusz Gajda^a ; ^{janusz.gajda@pwr.edu.pl}
• 关键词：Covariation ; αα-stable distribution ; Estimation ; Yule-Walker equations
• 刊名：Physica A: Statistical Mechanics and its Applications
• 出版年：2017
• 出版时间：1 March 2017
• 年：2017
• 卷：469
• 期：Complete
• 页码：588-603
• 全文大小：936 K
• 卷排序：469

文摘

We present measures of dependence adequate to alpha-stable processes. We propose an estimator of covariation measure for alpha-stable distributed random vectors. We propose a modified method of Yule-Walker to estimate the parameters of periodically autoregressive (PAR) time series model with alpha-stable innovations. We apply proposed method for PAR model to electricity market data describing the volume of up-regulating bids in Norway.