Abstract
Data envelopment analysis (DEA) is a nonparametric modelling technique which allows the user to model the relative efficiency of mutual funds based on defined inputs and outputs. The DEA analysis involves creating an efficient frontier using a data set and identifying inefficient mutual funds which need improvement. Recent literature has focussed on using performance metrics such as Jensen's alpha, Treynor measure and Sharpe ratio to measure performance of mutual funds. This thesis extends current literature in two important aspects. First, this study looks at the DEA efficiency of mutual funds in the United States based on defined inputs and outputs. Secondly, the study goes on to predict upgrades and downgrades of the mutual funds' Morningstar rating based on how DEA efficient the mutual fund is compared to its peers. This study investigates whether there is a relationship between DEA efficiency and Morningstar ratings up to six months in advance. In addition, this study investigates whether DEA efficient (inefficient) mutual funds receive an upgrade (downgrade) in Morningstar rating within the six months. Using this setup, the study finds that there is a significant positive association between DEA efficiency scores and Morningstar ratings. This study also finds significant results for predicting upgrades and downgrades in Morningstar ratings based on DEA efficiency up to six months following the calculation of efficiency scores. The benefit to investors will be the ability to calculate how efficient their mutual fund is compared to the other available mutual funds in the industry and buy or sell the mutual fund based on the results. This study will also benefit mutual fund managers as they will be able see how efficient their mutual fund is compared to their peers, which inputs and outputs need improvements and adjust accordingly. If used appropriately, this will also help mutual fund managers prevent possible job loss for poor performing mutual funds.