Dependence Measure of Daily versus Weekly Returns


Ab Razak, Ruzanna and Aminuddin, Nurul Hanis and Ismail, Noriszura (2018) Dependence Measure of Daily versus Weekly Returns. International Journal of Engineering and Technology(UAE), 7 (3.20). pp. 329-333. ISSN 2227-524X

[img] Text
document.pdf - Published Version
Restricted to Repository staff only

Download (274kB)


The copula method has been popular among researchers, especially in measuring the overall dependence and extreme dependence of multivariate data. Many copula studies have been focusing on examining the correlation of bivariate daily, monthly or weekly returns to explain the co-movement between financial markets and possible financial implications on portfolio management. Differently from past studies, this paper investigates whether different frequency of bivariate data (daily and weekly returns) possesses different dependence structures. The data from Kuala Lumpur Composite Index (KLCI) and Bursa Malaysia Hijrah Shariah Index (FBMHS) for the sample period of 2008 Q1 to 2017 Q1 are used for studying the dependency. The findings from this study reveal that both daily and weekly bivariate returns have the same dependence structure but different degree of dependence. Bivariate weekly returns showed stronger dependence compared to bivariate daily returns. This paper also highlights the statistical properties of weekly and daily data. The evidence from this research draws inferences for further study that lower frequency data such as monthly or quarterly returns data may have higher degree of dependence while higher frequency data may have lower degree of dependence and different copula structure.

Item Type: Article
Uncontrolled Keywords: Copulas (Mathematical statistics), Copula, KLCI, FBMHS
Subjects: Q Science > QA Mathematics > QA273-280 Probabilities. Mathematical statistics
Divisions: Faculty of Management (FOM)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 02 Dec 2020 17:33
Last Modified: 02 Dec 2020 17:33


Downloads per month over past year

View ItemEdit (login required)