Empirical Irregularities in the Estimation of Beta: The Impact of Alternative Estimation Assumptions and Procedures

Beta Estimation and Thin Trading: Evidence from Bahrain Bourse

“…They also noted that the variation in the estimates is often statistically significant and that estimates from weekly returns calculated on Monday or at the end of the month were significantly different from those derived for other days of the week or selected days of the month. The results reported by Asamoah and Quartey-Papafio (2011) lend further support to the findings of Draper and Paudyal (1995). They found that estimated betas are sensitive to the day-of-the-week effect.…”

“…However, they stop short of making any recommendation about which frequency should be used in estimating beta. Draper and Paudyal (1995) consider the other problem, the relationship between the length of the estimation period and beta estimates. They use a daily return interval over different estimation periods, and conclude that the best beta estimates can be obtained using around 400 observations.…”

“…Draper and Paudyal (1995) found that UK stock betas have day of the week or day of the month effects when using weekly or monthly observation, respectively. They also noted that the variation in the estimates is often statistically significant and that estimates from weekly returns calculated on Monday or at the end of the month were significantly different from those derived for other days of the week or selected days of the month.…”

Wealth effects to bidding companies from regulatory interventions in the UK

“…Draper and Paudyal (1995) examined the variability of estimates using the market model. They suggest that a large number of observations should be used to estimate and those methods of estimating should be carefully appraised before a particular model is selected (especially if the underlying returns data is not normally distributed).…”

Pot of gold or winner's curse? An event study of the auctions of 3G mobile telephone licences in the UK

“…11 While ordinary least squares (OLS) estimation remains widely used in event studies, accumulating evidence from diagnostic tests suggests frequent violation of the assumptions on which it is based. In particular, serious problems of nonnormality can arise due to both excess kurtosis (for example, Chan and Lakonishok (1992) and Mills, Coutts and Roberts (1996)) and also skewness (for example, Brown and Warner (1985), Chan and Lakonishok (1992), Campbell and Wasley (1993) and Draper and Paudyal (1995)). Moreover, since in the present study we are estimating abnormal returns for individual companies and small portfolios only, we cannot rely on the helpful aggregation effects which can restore normality in event-study test statistics when these are averaged over larger samples of companies (Cable and Holland, 2000).…”