For given change points \(\tau_1, \ldots, \tau_k\), compute the profile maximum likelihood estimates \(\lambda_1, \ldots, \lambda_{k + 1}\) for the values of the piecewise constant hazard function in a piecewise Exponential survival model. Standard errors for these estimates are provided as well, based on standard maximum (profile) maximum likelihood theory.
lambda_j_Exp(tau, time, event)A list containing the following elements:
The esimtates of the \(\lambda_j\).
The diagonal of the corresponding Hessian matrix. Note that the off-diagonal elements of the Hessian are all equal to 0.
Single number or vector with values of change points.
Event times, censored or observed, in months.
Censoring indicator, 1 for event, 0 for censored.
Kaspar Rufibach (maintainer)
kaspar.rufibach@roche.com
Fang, L., Zheng, S. (2011). A hybrid approach to predicting events in clinical trials with time-to-event outcomes. Contemp. Clin. Trials, 32, 755--759.
Goodman, M.S., Li, Y., Tiwari, R.C. (2011). Detecting multiple change points in piecewise constant hazard functions. J. Appl. Stat, 38(11), 2523--2532.