Citation

Aridi, Nor Azliana and Tan, Siow Hooi and Chin, Wen Cheong (2023) A stylized facts comparison between low-frequency and high-frequency data of Brazil stock market index. In: 5th ISM International Statistical Conference 2021: Statistics in the Spotlight: Navigating the New Norm, ISM 2021, 15-17 Aug 2021.

Full text not available from this repository.

Abstract

This study attempts to analyze short-term and long-term time series behavior of volatility by considering the volatility cascade of heterogeneity to generate the stylized facts which have distinctive parameters and attributes that is unavailable in low-frequency data. The information on the stylized facts is significant to improve the accuracy of the volatility forecast based on a different set of volatility model at a different time horizon. The data analysis involved the calculation of daily absolute returns and realized volatility calculated from the 5-min intraday returns of Brazil stock market (BOVESPA) over the period of 2014 - 2020. Both daily and 5-min returns have the kurtosis value lower than 3 which shows that both series are not experienced heavy-tailed. Based on the Box-Pierce Q-Statistics, both series displays the significant result of absence autocorrelations. The result also illustrates that both series features volatility clustering, volatility persistence and leverage effects. The Hurst exponent parameter based on R/S statistics indicates the existence of long-range dependence in high-frequency stock market only. The presence of jumps on the daily and intraday data has been detected by using test for additive jumps in ARMA-GJR (1,1) model and the result shows that more jumps are detected in high-frequency data compared to low-frequency data. All these findings are useful on further research to forecast stock market volatility by using high-frequency data due to the increasing information of stock market at intraday frequencies. The huge availability of high-frequency data has not only contributed to the improvement of previous standard volatility models but has also encouraged further research that not only consider short term period but also a longer horizon volatility forecast

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	stock market
Subjects:	H Social Sciences > HG Finance > HG4501-6051 Investment, capital formation, speculation > HG4551-4598 Stock exchanges
Divisions:	Faculty of Management (FOM)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	07 Apr 2023 01:53
Last Modified:	07 Apr 2023 01:53
URII:	http://shdl.mmu.edu.my/id/eprint/11297

Downloads

Edit (login required)