Abstract:
Using some ideas of statistical diagnostics, we analyse structuralchange in mathematical models of social economy from Bayesian predictive point of view. In this paper, we present two new methods which are based on the conditional predictive discordancy diagnostics and Kullback-Leibler divergence. Using these methods, we need no assumption about the structural change before analysing change points (in general methods, the assumption is mecessary, as a change in one or more of the parameters of the model in question). The methods presented are simple and generally acceptable and convenient for computing. Furthermore, we study relationhips between Chinese average steel consumption and average GNP. It is shown the results reflect well the changes of economy relationships by investigating the characteristics of government policies.