Improved beta estimation on the Johannesburg Stock Exchange: A simulation study

Abstract

In this article we focus on beta estimation in the thinly-traded environment of the Johannesburg Stock Exchange (JSE). We build on existing literature by evaluating a beta estimation procedure known as the trade-to-trade which has not until now been considered in the context of the JSE. We contrast our results with two known estimation procedures, i.e. the Cohen et al. and the traditional ordinary least squares (OLS). The trade-to-trade methodology, the estimator proposed by Cohen et al. and OLS are objectively assessed for shares typical of the JSE on the basis of unbiasedness and efficiency in the controlled environment of a simulation study. The trade-to-trade technique is found to be superior on both counts and is recommended as the appropriate technique for beta estimation on the JSE.

Estimation of betas of stocks with low liquidity

“…The SW method did not eliminate the problem of the downwards bias, when calculated with a daily periodicity, but did not present, for the average beta, a significant bias with weekly and monthly periodicity. This result is in line with that obtained by Bowie and Bradfield (1993), which, simulating monthly returns for the non-liquid series, show that the TT method produces a beta estimator superior to that obtained by the method of Cohen et. al.…”

“…(Brooks, 2001: Dimson (1979) simulated data according to eq. 3, Bowie and Bradfield (1993) simulated data according to eq. 6, and this paper simulated data according to eq.…”