Kalman filters have been used as a solution to engineering problems in the field of linear filtering and prediction for over twenty-five years. Kalman filters have also found applications in non-typical engineering areas. This dissertation examines the use of a Kalman filter to forecast intraday market prices;Several stock indexes and commodities are examined for autocorrelation patterns. These indicate that a market is not a random walk process. Markets exhibiting a statistically significant correlation pattern are modeled with Gauss-Markov, damped cosine, and ARIMA models. Each model is used by the Kalman filter to provide price forecasts. Minimum mean-square error is used as the performance index for model comparison;Buy and sell strategies are examined to determine if the Kalman filter forecasts can be utilized to provide increased profits. The buy and sell strategies are based on (1) speculator and (2) consumer viewpoints. Profits from the speculator strategies are compared against a buy-and-hold strategy and a slope-projection strategy. The consumer strategy uses the Kalman filter to reduce expenses for an operation which must continuously purchase goods from a market.