Estimating Systematic Risk – The Return Interval and Estimation Period Issues Evidence from Malaysia 2000-2006

• Main Authors: Ho, Catherine, Soke Fun, Chen, Yin Foo
• Format: Article
• Language: English
• Published: INTI International University 2009
• Subjects: H Social Sciences (General); HB Economic Theory
• Online Access: http://eprints.intimal.edu.my/1511/1/2009_p44.pdf; http://eprints.intimal.edu.my/1511/; https://intijournal.intimal.edu.my/intijournal.htm

Estimating the beta coefficient is central to the CAPM concept of rewarding the investors according to the systematic risk of an asset. However, while the concept is intuitively appealing, the estimation is biased by measurement issues such as thin trading, regression tendency, stability and choice of interval issues. While techniques have been developed to address the regression tendency, thin trading biases, no specific rules on the interval issues have been formulated. The trade-off between a longer estimation period for more observations and accuracy has to be weighted for a biased coefficient resulting from higher measurement errors. The results for this study provided support that daily returns provided the most efficient estimation in terms of smallest estimated coefficient errors but biased as any estimation period more than three years saw half of the sample experiencing a shift in their estimated beta.