The central limit theorem under censoring (Q5933572)

Let \(T_i\), \(i= 1,\dots, n\), be i.i.d. random variables with common distribution function \(F\). Put \(S= 1-F\). The observed data consist of \(X_i= \min(T_i, C_i)\) and \(\Delta_i= I(X_i= T_i)\), \(i= 1,\dots, n\), where \(C_1,\dots, C_n\) are i.i.d. random variables which are independent of \(T\)'s, and \(I(E)\) denotes the indicator of the event \(E\). Denote by \(\widehat S\) the Kaplan-Meier product-limit estimator of \(S\) based on the observed data. Let \(\varphi\) be any measurable function such that \(\int\varphi^2 dF< \infty\). The author obtains a central limit theorem for the Kaplan-Meier integral \(\int\varphi d\widehat F\), where \(\widehat F= 1-\widehat S\), and derives an i.i.d. representation for it.