nll_ditrunc_elife

Computes the log-likelihood for various parametric models suitable for threshold
exceedances. If threshold is non-zero, then only right-censored, observed event time and interval censored
data whose timing exceeds the thresholds are kept.

internal

A collection of parametric and nonparametric methods for the analysis of survival data. Parametric families implemented include Gompertz-Makeham, exponential and generalized Pareto models and extended models. The package includes an implementation of the nonparametric maximum likelihood estimator for arbitrary truncation and censoring pattern based on Turnbull (1976) <doi:10.1111/j.2517-6161.1976.tb01597.x>, along with graphical goodness-of-fit diagnostics. Parametric models for positive random variables and peaks over threshold models based on extreme value theory are described in Rootzén and Zholud (2017) <doi:10.1007/s10687-017-0305-5>; Belzile et al. (2021) <doi:10.1098/rsos.202097> and Belzile et al. (2022) <doi:10.1146/annurev-statistics-040120-025426>.

Leo Belzile

longevity

Statistical Methods for the Analysis of Excess Lifetimes

nll_ditrunc_elife function

<dl><dt>par</dt>
<dd>vector of parameters</dd>
<dt>time</dt>
<dd>excess time of the event of follow-up time, depending on the value of event</dd>
<dt>ltrunc1</dt>
<dd>lower truncation limit, default to <code>NULL</code></dd>
<dt>rtrunc1</dt>
<dd>upper truncation limit, default to <code>NULL</code></dd>
<dt>ltrunc2</dt>
<dd>lower truncation limit, default to <code>NULL</code></dd>
<dt>rtrunc2</dt>
<dd>upper truncation limit, default to <code>NULL</code></dd>
<dt>family</dt>
<dd>string; choice of parametric family, either exponential (<code>exp</code>), Weibull (<code>weibull</code>), generalized Pareto (<code>gp</code>), Gompertz (<code>gomp</code>), Gompertz-Makeham (<code>gompmake</code>) or extended generalized Pareto (<code>extgp</code>).</dd>
<dt>thresh</dt>
<dd>vector of thresholds</dd>
<dt>weights</dt>
<dd>weights for observations</dd>
<dt>arguments</dt>
<dd>a named list specifying default arguments of the function that are common to all <code>elife</code> calls</dd>
<dt>...</dt>
<dd>additional arguments for optimization, currently ignored.</dd></dl>

Arguments

Likelihood for doubly interval truncated data — nll_ditrunc_elife

<dl>

<dt>par</dt>
<dd>vector of parameters</dd>


<dt>time</dt>
<dd>excess time of the event of follow-up time, depending on the value of event</dd>


<dt>ltrunc1</dt>
<dd>lower truncation limit, default to <code>NULL</code></dd>


<dt>rtrunc1</dt>
<dd>upper truncation limit, default to <code>NULL</code></dd>


<dt>ltrunc2</dt>
<dd>lower truncation limit, default to <code>NULL</code></dd>


<dt>rtrunc2</dt>
<dd>upper truncation limit, default to <code>NULL</code></dd>


<dt>family</dt>
<dd>string; choice of parametric family, either exponential (<code>exp</code>), Weibull (<code>weibull</code>), generalized Pareto (<code>gp</code>), Gompertz (<code>gomp</code>), Gompertz-Makeham (<code>gompmake</code>) or extended generalized Pareto (<code>extgp</code>).</dd>


<dt>thresh</dt>
<dd>vector of thresholds</dd>


<dt>weights</dt>
<dd>weights for observations</dd>


<dt>arguments</dt>
<dd>a named list specifying default arguments of the function that are common to all <code>elife</code> calls</dd>


<dt>...</dt>
<dd>additional arguments for optimization, currently ignored.</dd>

</dl>

nll_ditrunc_elife: Likelihood for doubly interval truncated data

Description

Usage

Value

Arguments