scientific article; zbMATH DE number 4086755

From MaRDI portal
Publication:3814550





For analyzing failure time data, the author proposes a discrete time logistic regression model \[ P^{\beta}(\Delta N_ j(v)=1| {\mathcal G}_{v-1})=Y_ j(v-1)L(\beta 'Z_ j(v-1)),\quad v=1,2,... \] where \(Y_ j(v-1)=1\) if the j-th individual is at risk at time v and zero otherwise, \(\Delta N_ j(v)=1\) \((=0)\) if individual j at risk v fails (survives), \(Z_ j(v-1)\) is a vector of time-dependent covariates, L is the inverse of the logit function, and \({\mathcal G}_{v-1}\) contains the information at time v-1. The parameter of interest \(\beta\) is estimated via maximum-likelihood. Both consistency and asymptotic normality of the MLE are proved.











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 Q3814550)