Sequential estimation of survival functions with a neutral to the right process prior (Q1332486)

Let \(S\) be a survival function on the positive real line and let \(T_ i\) \((i=1,\dots, n)\) be a random sample from \(S\). The authors discuss a sequential estimation problem for \(S\) from a Bayesian viewpoint using a neutral to the right process prior (N.R.P.), and derive a sequential estimator based on the authors' 1-SPLA (1-stage-parametric-look-ahead) technique [Trab. Estad. 1, No. 1, 70-87 (1986; Zbl 0654.62037)]. They prove that the 1-SPLA rule is equal to the 1-SLA (1-stage-look-ahead) rule if N.R.P. is a Dirichlet prior. Furthermore, they derive asymptotically pointwise optimal rules in the sense of \textit{P. J. Bickel} and \textit{J. A. Yahav} [Proc. Fifth Berkeley Symp. Math. Stat. Probab., Univ. Calif. 1965/1966, 1, 401-413 (1967; Zbl 0214.451)].