Markov Model for Stock Market Buy and Sell Strategy

Abstract

Approximately 50% of the households in America invest in the stock market. In many cases the investor's buy and hold strategy leads to negative returns. Given the stock market fluctuates up and down as time progresses, analysts examine stock prices, volumes traded and ratios to recommend buy and sell opportunities.

The pattern of price and volume changes provides input for the analysts' recommendations. Using these patterns in a Markov model, this dissertation contains an intensive analysis of 41 securities over a 13-year period. The model establishes states of change in price and volume and calculates the best investor action for an individual security.

The selection of the proper security enhances the investor's probability of achieving an exceptional return. The research examines the correlation of price and volume characteristics to overall return. With the proper correlation, higher yielding securities may be selected. Once selected, the dissertation's research recommends when to switch from one security to another security. Also, periods of staying in a cash position are recommended.

Overall this model outperforms the average yearly buy and hold return of eleven percent by about four additional percentage points. Even subtracting the cost of the transactions, the model buys and sells the securities to obtain the additional return.