Sequential method for selecting the best probability model for equities

Abstract

If market-timing approach is used in investment strategy, the information on how market will perform in the next time period is needed. There are periods of time when equities perform much better than risk-free investment such as treasury bills. However, there are also periods when treasury bills perform better than equities. This thesis suggests a number of reasonable models for estimating the probability of equities outperforming treasury bills and for modeling the behavior of stock returns. Rosenbloom's sequential method (1999) is modified and used in order to determine which of the models is the best. This study has identified as a potential tool for selecting the best probability model for estimating the probability of equities outperforming treasury bill.