An Evaluation of Alternative Equity Indices - Part 2: Fundamental Weighting Schemes

Abstract

In this paper we explore an alternative approach for determining constituent weights for equity indices. This approach makes use of alternative definitions of company size, and is referred to as Fundamental Indexation (Arnott et al (2005)). Based upon a data set that comprises the largest 1,000 US stocks for each year in our sample, our results show that between 1968 and 2011 the fundamental index alternatives that we consider have out-performed a comparable index constructed on the basis of the market capitalisation of the index constituents in risk-adjusted terms. Our Monte Carlo experiments show that this superior risk-adjusted performance cannot be attributed easily to luck. We also find that although the superior performance is achieved with higher constituent turnover than required using the Market-cap approach to index construction, the turnover is lower, and in some cases much lower, than required by some of the heuristic and optimised index construction techniques that we explored in our last paper. Finally, we find that although the application of a simple market-timing rule does not enhance the returns on these fundamentally-weighted indices very significantly, it does reduce the volatility of their returns and their maximum drawdown quite considerably.