摘要
应用金融证券组合理论和Bayes方法构造一种证券收益预测模型.由多层Bayes的理论,对模型中含有的不确定参数给出超先验分布;并用Bayes方法确定超参数的分布.这种方法不仅使用了更多的信息不断修正先验,更重要的是允许模型中含有不确定参数,并自动考虑不确定参数估计的误差对分析结果的影响.
By Applying the financial securities combination theory and Bayes method a kind of securities profit prediction model is constructed. By the hierarchical Bayes theory, giving the exceed prior distribution to the uncertain parameters in the model, and using the Bayes method to determine the distribution of the exceed parameters this method not only uses more information and corrects priori continuously, more importantly, the presence of uncertain parameters in model are allowed, and automaticly considers the effects of the uncertain parameters estimation' s error to the analysis of the conclusion.
出处
《哈尔滨理工大学学报》
CAS
2007年第6期96-99,共4页
Journal of Harbin University of Science and Technology
基金
黑龙江省自然科学基金项目(G200521)