Remarks on Computations in Parametric and Semiparametric Estimation

Abstract

The literature on parametric and nonparametric estimation for models considered in the previous chapters is enormous. See, for example, Wu et al., Lifetime Data Anal, 8:21–34, (2002), Dabrowska, Sankhya, 67:153–187, (2005), Dabrowska, Second Eric L. Lehmann symposium—optimality, (2006), Bagdonavicius and Nikulin, Lifetime Data Anal, 5:323–344, (1999), Bagdonavicius and Nikulin, Accelerated life models, (2002), Bagdonavicius and Nikulin, Quantitative methods in cancer and human health risk assessment, (2005), Martinussen and Scheike, Dynamic regression models for survival functions, (2006), Scheike, Lifetime Data Anal, 12:461–480, (2006), Zeng and Lin, J R Stat Soc, B69:509–564, (2007), etc. Methods of estimation depend on experimental plans, censoring mechanism, covariate types, etc. Here we only give two general approaches, for parametric and semiparametric cases, which work well for all models. We have several remarks to help clarify the considered models.