利用次序截尾数据线性回归方法推断正态总体的未知参数

Using Order Censored Data Regression Method to Infer the Normal Overall Unknown Parameter

摘要: 统计学中，在采用抽样分析的方法进行参数估计与推断时，常要求样本数据数量、完整性好，且须满足随机抽取等特点。在实际应用中会遇到一系列按顺序排列的截尾样本数据，用普通的参数估计方法得到的参数估计值误差可能会较大，本文将回归分析方法推广到利用次序截尾数据推断正态总体参数中，且对所做结果进行模拟计算。通过误差比对，认为该方法是可行的，本文最后利用该方法通过部分排序的学生成绩推断了总体的平均成绩。

Abstract: The characteristics such as sample data amount, good integrity, and meeting the random drawing are often required in using the method of sampling analysis for parameter estimation and inference in statistics. However, in practice we are often faced with a series of order censored data, and the parameter error obtained by the usual estimation methods may be very big. In this paper, the method of regression analysis was used to deduce the overall parameters by using the order truncation data and to do the simulation calculation. It is shown that the method is feasible by comparing the error. Finally we used the method to solve the problem about estimating the overall average score by using some sort of student achievement.