Estimating equation methods for quantile regression with length-biased and right-censored data

Abstract

∑ i=1 (1 − ζ i)ξ i (τ) + o (ν,τU ] (1). 在给定 {A i , Y i , δ i , Z i } n i=1 的条件下, K * (τ) 弱收敛到一个均值为 0、 协方差矩阵Ω(s, t) = E{ξ i (s)ξ i (t) T } 的 Gauss 过程. 当给定的模型成立时, 对于 s, t ∈ [ν, τ U ],Ω(s, t) 几乎处处收敛到 Ω(s, t).

Quantile regression models for survival data with missing censoring indicators

“…Quantile regression 1 retains the direct interpretation as the accelerated failure time (AFT) model 2 and provides a natural way to model the data with heterogeneity; thus, it has become an appealing alternative to the Cox proportional hazards model 3 and the accelerated failure time model in survival analysis. 4–6 In the literature, a variety of approaches have been developed to handle the quantile regression model with right-censored survival data. Under the fixed censoring assumption, Powell 7,8 first proposed a regression quantile estimator for responses using the least absolute deviation (LAD) idea.…”