The goal of this study is to test the accuracy of
various mutual fund timing and selectivity models under a
range of portfolio managerial skills and varying market
conditions. Portfolio returns in a variety of skill
environments are generated using a simulation procedure. The
generated portfolio returns are based on the historical
patterns and time series behavior of a market portfolio proxy
and on a sample of mutual funds.
The proposed timing and selectivity portfolio returns
mimic the activities of actual mutual fund managers who
possess varying degrees of skill. Using the constructed
portfolio returns, various performance models are compared in
terms of their power to detect timing and selectivity
abilities, by means of an iterative simulation procedure.
The frequency of errors in rejecting the null hypotheses
of no market timing and no selectivity abilities shape the
analyses between the models for power comparison. The
results indicate that time varying beta models of Lockwood-
Kadiyala and Bhattacharya-Pfleiderer rank highest in tests of
both market timing and selectivity. The Jensen performance
model achieves the best results in selectivity environments
in which managers do not possess timing skill. The
Henriksson-Merton model performs most highly in tests of
market timing in which managers lack timing skill.
The study also investigates the effects of
heteroskedasticity on the performance models. The results of
analysis before and after model correction for nonconstant
error term variance (heteroskedasticity) for specific
performance methodologies do not follow a consistent pattern. / Graduation date: 1992
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/36490 |
| Date | 17 April 1992 |
| Creators | Azimi-Zonooz, Aydeen |
| Contributors | McDowell, Edward D. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0024 seconds