| Article ID: | iaor1990787 |
| Country: | United States |
| Volume: | 6 |
| Start Page Number: | 1 |
| End Page Number: | 7 |
| Publication Date: | Mar 1990 |
| Journal: | Communications in Statistics - Stochastic Models |
| Authors: | Mandarino J.V. . |
A model for trends in security prices proposed by S.J. Taylor is used to develop a trading strategy that maximizes the expected profit per epoch given an arbitrary convex cost function. The optimal strategy is found explicitly with the aid of a trend estimate based on the Kalman filter. Expressions are given for the optimal expected profit per epoch and for the expected profit per epoch when the trend is estimated using an arbitrary absolutely summable linear filter.