Dual Long Memory Properties with Skewed and Fat-Tail Distribution

Sang-Hoon Kang ； Seong-Min Yoon

《International Journal of Business and Information》 7卷2期 (2012/12) Pp. 225-249

https://doi.org/10.6702/ijbi.2012.7.2.4

並列摘要

This paper examines the long memory properties in both the returns and volatility of Korean stock prices. For this purpose, the ARFIMA-FIGARCH model was applied to the daily KOSPI and KOSDAQ return series. In the data analysis, the ARFIMA-FIGARCH model establishes the robustness of long memory results, although the presence of long memory is questionable in the returns of two daily indices. In addition, the assumption of non-normality is appropriate for capturing the asymmetry and tail fatness of estimated residuals. These findings suggest that the model based on the Gaussian normality assumption may be inappropriate for modeling the long memory property. Finally, the presence of long memory in the Korean stock market is not spurious as a result of market structural changes.

並列關鍵字

Asymmetry ； fat-tail ； long memory ； skewed student's t-distribution ； Korean stock market

參考文獻

Anderson, T.G.,Bollerslev, T.(1997).Heterogeneous information arrivals and return volatility dynamics: Uncovering the long-run in high frequency returns.Journal of Finance.52,975-1005.

Baillie, R.T.(1996).Long memory processes and fractional integration in econometrics.Journal of Econometrics.73,5-59.

Baillie, R.T.,Bollerslev, T.(1989).The message in daily exchange rates: A conditional-variance tale.Journal of Business and Economic Statistics.7,297-305.

Baillie, R.T.,Bollerslev, T.,Mikkelsen, H.O.(1996).Fractionally integrated generalized autoregressive conditional heteroskedasticity.Journal of Econometrics.74,3-30.

Beine, M.,Laurent, S.(2003).Central bank interventions and jmps in double long memory models of daily exchange rates.Journal of Empirical Finance.10,641-660.