考虑随机右删失数据下非线性回归模型,提出了模型中未知参数的调整的经验对数似然比统计量.在一定的条件下,证明了所提出的的统计量具有渐近x2分布,由此结果构造了兴趣参数的置信域.通过模拟研究,对经典的经验似然、调整的经验似然和非线性最小二乘方法在有限样本下进行了比较,并对氯离子浓度试验数据进行了分析.
In this paper, the nonlinear regression models with right censored data are considered, an ajusted empirical log-likelihood ratio is proposed for the vector of regression parameter. Under certain conditions, it is shown that the proposed statistics have the chisquared limiting distribution, which enables us to construct the confidence regions of the parameters of interest. In the simulation study, we have compared the performance of the finite sample for the classical empirical likelihood, the adjusted empirical likelihood and the nonlinear least squares methods. An illustration is given by using chloride transport data.