Extension of sharpe's asset class factor model : using ETFs to replicate returns of U.S. balanced mutual funds.

Abstract

Investors are becoming more financially sophisticated and they have poor perception of active fund management. We believe they would increasingly seek to manage their own portfolios. This is made easier with enhanced access to online brokerages and simple software like Microsoft Excel. Sharpe‟s Asset Allocation paper (1992) illustrated the application of the asset class factor model to analyze the performance of a set of open-end mutual funds between 1985 and 1989. In our paper, we replaced the asset classes‟ indexes with actively traded ETFs and conducted our study across two time periods of 2001 to 2010 and 2006 to 2010. Our results show that Style Long portfolio for 2006 to 2010 achieved the highest Style of 86.1%, comparable to Sharpe‟s results for balanced mutual funds of 89.0%. Style Long portfolios have highest positive tracking error. Retail investors can thus use this portfolio type to replicate the returns of top performing U.S. balanced funds. Empirical evidence from Fama and French’s APT study in 1993 also showed that factors like value and size were significant in explaining cross-sectional returns of stocks. The Style Long portfolio for 2006 to 2010 with highest R2 value was most useful in analyzing the asset exposures of mutual funds. Our study provides practical applications for retail investors who wish to manage their own portfolios using Style analysis.