On Tests for Self-exciting Threshold Autoregressive-type Non-linearity in Partially Observed Time Series

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Pages (from-to)43-62
Journal / PublicationJournal of the Royal Statistical Society. Series C: Applied Statistics
Volume40
Issue number1
Publication statusPublished - 1991

Abstract

We have adapted and extended the Petruccelli-Davies test, Petruccelli's test and Tsay's test for non-linearity in time series to cope with partially observed series. The Kalman filtering algorithm is used in the estimation stage to realize the adaptation. Two of the adapted tests are checked with a Monte Carlo study and all three tests are applied to three real series from the financial world. The fine tuning achieved by allowing for closing date effects offers further insights.

Research Area(s)

  • Hang Seng index, IBM stock price, Kalman filter, Non-linearity, Partial observations, Petruccelli-Davies test, Petruccelli's test, Self-exciting threshold autoregression, Share price, Tsay's test

Bibliographic Note

Research Unit(s) information for this publication is provided by the author(s) concerned.