DocumentCode :
3414333
Title :
Efficient estimation of the Hurst parameter in high frequency financial data with seasonalities using wavelets
Author :
Bayraktar, Erhan ; Poor, H. Vincent ; Sircar, K. Ronnie
Author_Institution :
Princeton Univ., NJ, USA
fYear :
2003
fDate :
20-23 March 2003
Firstpage :
309
Lastpage :
316
Abstract :
S&P 500 Index data taken at one-minute intervals over the course of 11.5 years (January 1989- May 2000) is analyzed, and in particular the Hurst parameter over segments of stationarity (the time period over which the Hurst parameter is almost constant) is estimated. (The segments of stationarity are a byproduct of our analysis, no prior assumption about it is made.) An asymptotically efficient estimator using the log-scale spectrum is employed. This estimator is robust to additive non-stationarities, and it is shown to be robust to multiplicative non-stationarities, i.e. seasonalities, as well. Analyzing cumulative sums of returns, rather than the returns themselves, is essential in removing the effect of seasonalities. It is shown that it is necessary to use wavelets with at least two vanishing moments for the analysis in order to achieve this robustness. This analysis shows that the market has become more efficient since 1997.
Keywords :
Brownian motion; econophysics; stock markets; wavelet transforms; Hurst parameter; S&P 500; fractional Brownian motion; log-scale spectrum; seasonalities; stationarity; wavelets; Books; Brownian motion; Cotton; Data engineering; Frequency estimation; Investments; Operations research; Robustness; Stochastic processes; Wavelet analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence for Financial Engineering, 2003. Proceedings. 2003 IEEE International Conference on
Print_ISBN :
0-7803-7654-4
Type :
conf
DOI :
10.1109/CIFER.2003.1196276
Filename :
1196276
Link To Document :
بازگشت