“…For the case that the response variable X in (1.2) is subject to interval censoring Case 1 or Case 2 as described in Groeneboom and Wellner (1992), there have been the works on the estimation of b 0 by Finkelstein (1986), Satten (1996), Huang (1997), among others. However, the approaches developed for right censored data and interval censored data do not have direct extension to other complicated types of censored data, such as doubly censored data (Turnbull, 1974;Chang and Yang, 1987;Gu and Zhang, 1993;Ren and Gu, 1997), partly interval-censored data (Huang, 1999), bivariate right censored data (Dabrowska, 1989), bivariate data under univariate right censoring (Lin and Ying, 1993), etc., for which there have not been any published works on the estimation of b 0 . Also, for these complicated types of censored data as well as for above mentioned interval censored Case 1 or Case 2 data, there have been no works on the testing methods to assess the goodness-of-fit for the Cox model (1.1).…”