piecewiseExp_profile_loglik_tau

In a piecewise Exponential survival model, the estimates for the value of the hazard function can be given explicitly, see Fang and Su (2011), and can be computed using <code>lambda_j_Exp</code>. The values of the change points can then be computed using the profile log-likelihood function. The function <code>piecewiseExp_profile_loglik_tau</code> computes the value of this profile log-likelihood function and can be used together with <code><a href="/link/optim?package=eventTrack&version=1.0.4" data-mini-rdoc="eventTrack::optim">optim</a></code> to compute the change points \(\tau_1, \ldots, \tau_{k}\).

htest

Implements the hybrid framework for event prediction described in Fang & Zheng (2011, <doi:10.1016/j.cct.2011.05.013>). To estimate the survival function the event prediction is based on, a piecewise exponential hazard function is fit to the time-to-event data to infer the potential change points. Prior to the last identified change point, the survival function is estimated using Kaplan-Meier, and the tail after the change point is fit using piecewise exponential.

Kaspar Rufibach

eventTrack

Event Prediction for Time-to-Event Endpoints

piecewiseExp_profile_loglik_tau function

<dl> <dt>tau</dt>
<dd>Single number or vector with values of change points.</dd> <dt>time</dt>
<dd>Event times, censored or observed, in months.</dd> <dt>event</dt>
<dd>Censoring indicator, 1 for event, 0 for censored.</dd></dl>

Arguments

Kaspar Rufibach (maintainer) <a href="/link/kaspar.rufibach%40roche.com?package=eventTrack&version=1.0.4" data-mini-rdoc="eventTrack::kaspar.rufibach@roche.com">kaspar.rufibach@roche.com</a>

Author

In a piecewise Exponential survival model, the estimates for the value of the hazard function can be given explicitly, see Fang and Su (2011), and can be computed using <code>lambda_j_Exp</code>. The values of the change points can then be computed using the profile log-likelihood function. The function <code>piecewiseExp_profile_loglik_tau</code> computes the value of this profile log-likelihood function and can be used together with <code><a href='https://rdrr.io/r/stats/optim.html'>optim</a></code> to compute the change points \(\tau_1, \ldots, \tau_{k}\).

Profile maximum log-likelihood function for change points in piecewise Exponential survival model — piecewiseExp_profile_loglik_tau

<dl>

 <dt>tau</dt>
<dd>Single number or vector with values of change points.</dd>

 <dt>time</dt>
<dd>Event times, censored or observed, in months.</dd>

 <dt>event</dt>
<dd>Censoring indicator, 1 for event, 0 for censored.</dd>

</dl>

Kaspar Rufibach (maintainer) <a href='mailto:kaspar.rufibach@roche.com'>kaspar.rufibach@roche.com</a>

piecewiseExp_profile_loglik_tau: Profile maximum log-likelihood function for change points in piecewise Exponential survival model

Description

Usage

Value

Arguments

Author

References