The purpose of this paper is to investigate the preference change in the demand for meat subject to random coefficients in Saudi Arabia. A Fortran 77 program has been designed to estimate the demand function for meat using Kalman filtering techniques and maximum likelihood approach. The initial values of the coefficient and covariance estimates are an essential prior information in the Kalman filtering techniques. Results provide substantial random coefficients in red meat, implying important structural change occurs in red meat more than poultry and fish demand.