摘要
文章通过分析线性均值回归模型中随机误差的分位数,将线性均值回归关系式恒等变换为线性分位数回归关系式,把线性均值回归模型与线性分位数回归模型的研究方法整合起来,得到了依分位数水平的线性均值回归模型。建立了因变量取值中心与分位数之间依数据信息的联系,在同一分位数水平下,同时给出均值预测与均值分位数预测。
By analyzing the quantile of random error in linear mean regression model, this paper transforms the linear mean regression relation into linear quantile regression relation, and then integrates the linear mean regression model and linear quantile regression model to obtain a linear mean regression model at quantile level. Finally, the paper establishes the relationship between the value center of the dependent variable and the quantile according to the data information, and presents the mean prediction and mean quantile prediction at the same quantile level.
作者
刘兆君
Liu Zhaojun(School of Mathematics and Information Science,Shandong Technology and Business University,Yantai Shandong 264005,China)
出处
《统计与决策》
CSSCI
北大核心
2022年第11期56-59,共4页
Statistics & Decision
作者简介
刘兆君(1965-),男,山东烟台人,教授,研究方向:不确定性的数学理论。
