The impact of the return interval on the estimation of systematic risk

Timothy J. Brailsford*, Thomas Josev

*Corresponding author for this work

Research output: Contribution to journalArticleResearchpeer-review

24 Citations (Scopus)


The estimation of systematic risk (or 'beta') is central to the implementation of the capital asset pricing model and the market model for both researchers and practitioners. It is well known that a variety of beta estimates can result for the one stock depending on various factors such as the calculation of returns, choice of the market index, sample period and length of the estimation period. In this paper, we are concerned with one such factor being the interval over which returns are measured. The impact of the return interval on the beta estimate is known as the 'interval effect'. There is only limited evidence on the impact of the interval effect outside the US equity market. As such, this paper first documents the impact of the effect in the Australian equity market. The initial results indicate that the beta estimates of high (low) capitalised firms fall (rise) as the return interval is lengthened. The paper then provides an understanding of the effect by testing the model proposed by [Hawawini, G., 1983. Why beta shifts as the return interval changes, Financial Analysts Journal 39, 73-77]. This model provides a prediction of the size and direction of change in the beta estimate as a result of changes in the return interval. The empirical results generally support the predictions. These findings have implications for the use of beta estimates in portfolio and risk management, measurement of abnormal returns and testing of asset pricing models.

Original languageEnglish
Pages (from-to)357-376
Number of pages20
JournalPacific Basin Finance Journal
Issue number3
Publication statusPublished - Jul 1997
Externally publishedYes


Dive into the research topics of 'The impact of the return interval on the estimation of systematic risk'. Together they form a unique fingerprint.

Cite this