Abstract
The theory of so-called ‘optimal’ portfolio rebalancing has emerged over the last decade in journals and working papers, but this theory has several drawbacks, being based on advanced mathematics and relying on sets of restrictive assumptions. This paper investigates rebalancing strategies by using resampling procedures from actual South African data in a way that captures the intrinsic high volatility nature of South African asset price movements. It considers the main consequences of calendar and range rebalancing strategies on tracking error, transaction costs and portfolio performance and demonstrates that range rebalancing has advantages over calendar rebalancing in the South African financial context using the comparative portfolio allocations of Firer, Peagram and Brunyee (2003). As such it provides a practical framework for South African portfolio managers to make informed choices on the appropriate approach for best-practice portfolio rebalancing.