ena_051705_165248 - The performance of alternative...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
The performance of alternative techniques for estimating equity betas of Australian firms REPORT PREPARED FOR THE ENERGY NETWORKS ASSOCIATION May 2005 Prepared by: Stephen Gray, University of Queensland Business School Jason Hall, University of Queensland Business School Jerry Bowman, University of Auckland Tim Brailsford, University of Queensland Business School Robert Faff, Monash University Bob Officer, Capital Research
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
i | May 2005 | Contents Executive summary. ...................................................................................... 1 1 Introduction . ........................................................................................ 2 2 Motivations for different estimation methods . .................................... 4 2.1 Changing the Data Estimation Period. ..................................................... 4 2.2 The use of industry betas rather than company-specific betas . ............ 6 2.3 Blume-adjusted betas. ............................................................................... 10 2.4 The default beta or null hypothesis of unity. ........................................ 11 2.5 The removal of outlying observations. .................................................. 13 2.6 The removal of unusual market events . ................................................ 15 3 How the ‘Best’ Estimation Method is Determined. .......................... 18 4 Empirical Methods and Results. ........................................................ 22 4.1 Optimal estimation period. ...................................................................... 23 4.2 Blume Beta. ................................................................................................ 25 4.3 The default beta or null hypothesis of unity. ........................................ 26 4.4 Industry Beta . ............................................................................................ 29 4.5 Outlier-adjusted beta estimates. .............................................................. 31 4.6 Unusual market events. ............................................................................ 34 5 Robustness of the Results. ................................................................. 37 5.1 Restricting to a Constant Sample . .......................................................... 37 5.2 Increasing the Length of the Forecast Period . ..................................... 37 5.3 An Examination of the Utility Industry . ............................................... 37 5.4 Results for Extreme Betas. ...................................................................... 38 5.5 Reducing the Data Restrictions in the Beta Estimation. ..................... 38 5.6 Other Error Metrics and Further Tests of Significance. ..................... 38 5.7 Analysis of the Economic Significance of the Results. ....................... 39 5.8 Other Beta Adjustments and Estimation Methods . ............................ 39 6 Conclusions and Recommendations . ................................................ 40
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
1 | May 2005 | Executive summary Executive summary We evaluate the ability of estimated equity betas to explain observed equity returns of Australian-listed stocks from 1989 to 2003 under alternative estimation techniques. If equity beta estimates are to be used in the CAPM to estimate the cost of equity capital, the optimal technique is that which provides the best match between observed stock and market returns. That is, the beta estimation technique that is chosen is that which “works best” when used in the CAPM to model equity returns. The standard approach taken by Australian regulators, who perform comparable-company analysis using firm- specific beta estimates, is to rely on beta estimates from an ordinary least squares (OLS) regression on 48 months of share market and company returns. The majority of this paper examines how best to estimate the equity beta of each individual firm. Subsequent work will investigate more thoroughly how best to aggregate estimates from comparable firms. Our analysis indicates that the standard OLS beta estimates based on four years of monthly data perform worse than almost every alternative that we examine. That is, the mechanical beta estimates that are produced by data service providers perform poorly relative to other estimation techniques – they simply do not work as well as other techniques when used in the CAPM to model equity returns. In particular, we find that this technique is inferior to the following alternatives:
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 49

ena_051705_165248 - The performance of alternative...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online