riskdist

Make a plot of predicted risks by outcome

internal

Fit calibrations curves for clinical prediction models and calculate several associated
metrics (Eavg, E50, E90, Emax). Ideally predicted probabilities from a prediction model
should align with observed probabilities. Calibration curves relate predicted probabilities
(or a transformation thereof) to observed outcomes via a flexible non-linear smoothing function.
'pmcalibration' allows users to choose between several smoothers (regression splines, generalized
additive models/GAMs, lowess, loess). Both binary and time-to-event outcomes are supported.
See Van Calster et al. (2016) <doi:10.1016/j.jclinepi.2015.12.005>;
Austin and Steyerberg (2019) <doi:10.1002/sim.8281>;
Austin et al. (2020) <doi:10.1002/sim.8570>.

Stephen Rhodes

pmcalibration

Calibration Curves for Clinical Prediction Models

riskdist function

<dl><dt>y</dt>
<dd>vector of binary outcome</dd>
<dt>p</dt>
<dd>vector of predicted risks</dd>
<dt>ypos</dt>
<dd>where to center the y axis</dd>
<dt>labels</dt>
<dd>labels for outcomes 0 and 1, respectively. Default to "0" and "1"</dd>
<dt>nbins</dt>
<dd>Default 101</dd>
<dt>add</dt>
<dd>if TRUE (default) added to an existing plot. If FALSE a new plot is made</dd>
<dt>maxh</dt>
<dd>maximum height of a bar (the bin with largest number of observations). Default = .15</dd></dl>

Arguments

Make a plot of predicted risks by outcome — riskdist

<dl>

<dt>y</dt>
<dd>vector of binary outcome</dd>


<dt>p</dt>
<dd>vector of predicted risks</dd>


<dt>ypos</dt>
<dd>where to center the y axis</dd>


<dt>labels</dt>
<dd>labels for outcomes 0 and 1, respectively. Default to "0" and "1"</dd>


<dt>nbins</dt>
<dd>Default 101</dd>


<dt>add</dt>
<dd>if TRUE (default) added to an existing plot. If FALSE a new plot is made</dd>


<dt>maxh</dt>
<dd>maximum height of a bar (the bin with largest number of observations). Default = .15</dd>

</dl>

riskdist: Make a plot of predicted risks by outcome

Description

Usage

Value

Arguments