It is well known that the CAPM beta is not stable over time. We investigate
the nature of the time-variation in betas using monthly Australian data fr
om 1979 to 1994 for 23 sectors. We discuss beta estimates for sub-periods a
nd tests of the statistical adequacy of the market model used to estimate t
he betas. We estimate time-varying betas using recursive regressions, rolli
ng regressions and using the Kalman Filter. We find considerable time-varia
tion in the estimated betas and find that many are non-stationary. We estim
ate a simple model which explains the variation in each of the betas in ter
ms of a time trend, allowing for a break both in level and in trend at Octo
ber 1987. The model explains a large proportion of the variation in the bet
as over the sample period for most of the sectors. (C) 1999 IMACS/Elsevier
Science B.V. All rights reserved.