Portfolio performance measures: Some extensions

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Business Administration


Ravi K. Shukla


Portfolio performance, Mutual funds, Return, Category benchmarks

Subject Categories

Business Administration, Management, and Operations | Portfolio and Security Analysis


By 1998, over three trillion dollars were invested in mutual funds. A fair and accurate measurement of the performance of portfolio managers is essential to determine if the managers are adding value to the investors' portfolios. There are basically two abilities that portfolio managers have that can be evaluated: their ability to select investments and the timing of their investment choices. Three new performance measures are developed in this dissertation: a selection measure, a historical return-based timing measure, and a benchmark return-based timing measure. The selection measure and the benchmark return-based timing measure use a benchmark of other investments that the manager could have invested in as a reference point. The historical return-based timing measure is referenced against the investments' last period's return. The new measures are constructed without the use of a market portfolio, asset pricing model, or statistical characteristics. The benchmarks that are used in the selection measure and the benchmark return-based timing measures are security-specific benchmarks rather than a market index that has been employed in past portfolio performance measures. Past portfolio performance measures have assessed the performance of a manager by comparing his/her portfolio to an index or general benchmark for the portfolio. The past measures are adjusted for risk by using some statistical measure, typically the beta or correlation of the manager's portfolio to the benchmark. The new measures also account for changes in risk in the portfolio that a manager may employ. These measures can be used to evaluate the manager's performance on different levels, as well as to assess different types of portfolios. In this study, an empirical evaluation of all equity mutual funds are conducted for a three-year and a six-year sample. We construct and analyze the category-based benchmarks using the three-digit standard industry code's. The data shows that the three digit standard industry code category-based benchmarks perform the same as or better than other common benchmarks in representing the risk return relationship for equity securities. We find that on average mutual fund managers have investment selection ability, but do not have the ability to time their investments. We also note from the analyses that mutual fund managers that have a higher (lower) turnover ratio have a lower (higher) timing ability, as well as being invested in more risky investments as measured by the beta measurement and standard deviations of their holdings. Additionally, we discover that there are some trends over time in the descriptive statistics, as well as in some of the measures. Assessment the samples based on their stated objectives show that the small company mutual funds have the worst timing ability, but have the highest turnover ratio and second highest expense ratio. Assessing the persistence of performance for the different measures indicate that persistence depends on the performance measurement technique applied.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.