compete.simulation

Simulate from a competing risk model with correlated log normal errors, and plot various estimates.

Proportional hazards estimation in the presence of a partially monotone likelihood has difficulties, in that finite estimators do not exist.  These difficulties are related to those arising from logistic and multinomial regression.  References for methods are given in the separate function documents.  Supported by grant NSF DMS 1712839.

John E Kolassa

PHInfiniteEstimates

Tools for Inference in the Presence of a Monotone Likelihood

compete.simulation function

<dl><dt>ncr</dt>
<dd>Number of competing risks.</dd>
<dt>sig</dt>
<dd>correlation among competing risks.</dd>
<dt>ns</dt>
<dd>number of observations.</dd></dl>

Arguments

Simulate from a competing risk model with correlated log normal errors, and plot various estimates. — compete.simulation

<dl>

<dt>ncr</dt>
<dd>Number of competing risks.</dd>


<dt>sig</dt>
<dd>correlation among competing risks.</dd>


<dt>ns</dt>
<dd>number of observations.</dd>

</dl>

compete.simulation: Simulate from a competing risk model with correlated log normal errors, and plot various estimates.

Description

Usage

Arguments