University of Southampton
Nonlinear Dynamics and Predictability of Japanese Stock Market Returns
Email address: firstname.lastname@example.org
There is now considerable evidence indicating that stock market returns may be predictable, and identifying variables relevant in predicting returns. Relatively little of this evidence relates to Japanese stock markets, which are often regarded as having characteristics that are different from those in other major stock markets (see, e.g., the discussion in Sentana and Wadhwani (1991)). We analyse the predictability of Japanese stock returns allowing for potentially nonlinear dynamics. Hamiltonís (2001) parametric approach to flexible nonlinear inference is used to conduct robust tests for nonlinearities in Japanese stock returns, and to suggest a relatively parsimonious nonlinear model that characterises effectively the predictability of Japanese stock market returns over the period 1970-1999. This estimated model sheds some light on the different characteristics of stock markets in Japan.
PDF file of paper: Not available.
Session: Empirical Finance
Time: Sunday, 8 July, 2:15pm - 3:45pm