ETF performance measurement — Data Envelopment Analysis

In this paper we apply Data Envelopment Analysis (DEA) to evaluate the performance of IShare World exchange-traded funds (ETFs) including major countries in Asia Pacific, Europe, North America and some emerging markets from 2006 to 2009. We investigate how the recent financial crisis affects the economies of different countries by looking into particular country indices relative to S&P 500 and Sharpe Ratio. Contrary to other DEA studies that rely on CCR and BCC models, an advanced model - range directional measure (RDM) - is employed in our work to measure ETF performance with an aim to overcoming the problem of negative-valued data. Another innovative approach is that we separate the total risk commonly measured by standard deviation into two components: upside potential and downside deviation. Finally, Spearman correlation coefficient is calculated to test whether the performances of ETFs are persistent over two consecutive time periods. The results indicate that DEA is able to provide a more comprehensive picture of ETF performance appraisal and hence a viable complement to traditional risk-adjusted index.