scientific article; zbMATH DE number 3928089

From MaRDI portal
Publication:3702277





[For the entire collection see Zbl 0561.00017.] The note addresses to random censoring, where only \(X_ i=\min (X^ 0_ i,Y_ i)\) and \(1_{[X^ 0_ i\leq Y_ i]}\) are observable. Interest is in estimation of parameters of the distribution function \(F^ 0\) of \(X^ 0_ i\) where no prior information about the distribution function of the censoring variables \(Y_ i\) is necessary. Four estimation methods are investigated: method of moments, the integrated squared error estimator, quantile estimators of scale and shift parameters and integral functionals of product limit. Their asymptotic properties are investigated by the way of studying weak convergence of parameter estimated processes. Also composite goodness-of- fit tests are dealt with for testing that the distribution function of \(X^ 0_ i\) belongs to a special parametric family.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3702277)