bestbeta

This function implements the approximate conditional inferential approach of kz19;textualPHInfiniteEstimates to proportional hazards regression.

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

bestbeta function

<dl><dt>fit</dt>
<dd>Output from a Cox PH regression, with x=TRUE and y=TRUE</dd>
<dt>exclude</dt>
<dd>data set with stratum and patient number to exclude.</dd>
<dt>start</dt>
<dd>Starting value</dd>
<dt>touse</dt>
<dd>columns of the design matrix to use.</dd>
<dt>usecc</dt>
<dd>Logical variable indicating whether to use a continuity correction, or nuerical variable representing teh continuity correction.</dd></dl>

Arguments

Newton Raphson Fitter for partial likelihood — bestbeta

<dl>

<dt>fit</dt>
<dd>Output from a Cox PH regression, with x=TRUE and y=TRUE</dd>


<dt>exclude</dt>
<dd>data set with stratum and patient number to exclude.</dd>


<dt>start</dt>
<dd>Starting value</dd>


<dt>touse</dt>
<dd>columns of the design matrix to use.</dd>


<dt>usecc</dt>
<dd>Logical variable indicating whether to use a continuity correction, or nuerical variable representing teh continuity correction.</dd>

</dl>

bestbeta: Newton Raphson Fitter for partial likelihood

Description

Usage

Value

Arguments

References

Examples