ABSTRACT
This chapter discusses the estimation of the means and variances of purely random coefficient and sequentially varying coefficient models, adaptive coefficients and stochastically convergent. It explains the problem of estimation of the mean coefficients and the variances. The chapter also discusses the estimation of the randomly varying coefficient models under two alternative specifications that is Adaptive Coefficient; and Stochastically Convergent Coefficient. Under both the chapter first considers the case where only the intercept is changing. Then it considers the situation where both the intercept and the coefficient of x are changing. The chapter focuses on two methods of estimating the means and variances of the purely random coefficients model, such as Classical Method: Generalised Least Squares (GLS) estimator, Maximum Likelihood (ML) estimator, and Bayesian Method. The Bayesian estimator of any parameter is the mean or mode of its posterior density function. When both coefficients follow a stochastically convergent variation the ML estimation technique can be similarly formulated.
