Estimation and Prediction in Mixture Truncated Normal Distribution

This paper is concerned with the estimation of the parameters of the mixture normal distribution with truncated components using Bayesian approach. One of numerical methods is used to obtain the joint posterior distribution of the variances of the two components, Predictive distribution of the future observation, is also obtained which is a mixture of two components. Each component is weighted by the belongance ratio of future observation to each component of mixture truncated normal distribution.