Pure quantile portfolios on the Johannesburg stock exchange

Abstract

Rules-based portfolio sorts are commonplace for the evaluation of styleanomalies. An unfortunate consequence of constructing portfolios on a target styleis the unintended loading on non-target factors. A plausible approach is the appli-cation of optimisation to maintain target factor loading while minimising non-targetfactor exposures. We test this methodology on an emerging market bourse, theJohannesburg Stock Exchange, via quintile portfolios sorted on momentum, valueand size. We find that value and momentum benefit most from optimisation interms of nominal and risk-adjusted performance. From an emerging market per-spective, we show that optimisation is a viable alternative when independent sorts are infeasible.