非线性模型在右删失数据下的统计诊断

Statistical Diagnostics for Nonlinear Models with Right-Censored Data

摘要: 本文研究了非线性模型在带有右删失数据下如何统计诊断的问题。首先用极大似然估计的方法求出了参数的估计问题。再对模型运用数据删除思想进行考量，求出了删除前后参数的公式，给出了判定影响点和异常点的一些统计量。最后，通过数值模拟分析，证明了该方法的可行性。

Abstract: This paper considers how to solve statistical diagnosis problem of the nonlinear models with right-censored data. First, we use the method of maximum likelihood estimates to reach the parameters. Based on the idea of case-deletion models, we obtain the formula of the parameters and propose some diagnostic statistics to determine outliers or influential points. Finally, we use numerical simulation analysis to verify the feasibility of the theory.