项目名称: 相依样本下的经验似然推断
项目编号: No.11271088
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 数理科学和化学
项目作者: 秦永松
作者单位: 广西师范大学
项目金额: 68万元
中文摘要: 在可靠性分析、医学试验、金融数据分析及时间序列分析等众多应用领域都有大量相依样本存在, 本项目系统研究一些常见相依样本(如负相协样本、正相协样本、混合样本、线性过程等)结构下的一类重要统计模型(如线性模型、非参数模型、半参数模型及广义线性模型)参数的经验似然置信区间的构造,得到一批在统计领域内具有国内外先进水平的理论成果,同时将上述理论应用到经济指标的预测,得到一些主要经济指标预测的新方法,为政府和有关部门决策提供科学依据。
中文关键词: 相协样本;经验似然;非参数模型;;
英文摘要: Dependent samples occour in many application fields such as reliability theory, clinical trial studies, financail data analysis and time series analysis. This project focuses on the constuction of empirical likelihood confidence intervals for parameters in a class of important statistical models such as linear regression models, nonparametric regression models, semi-parametric regression models and generlized linear models under a class of commonly encountered dependent data structures such as negetively associated smaples, positively associated samples, mixing samples and linear processes. We aim to o btain a series of excellent research outputs in the related statistical fields in the world. At the same time, we will use the theoritical results obtained to the prediction of economic measurements. We intend to obtian some new methods in the prediction of economic measurements, which could be used by goverment and related organizations for scientific decision-making.
英文关键词: associated sample;empirical likelihood;nonparametric model;;