predict.dpGLM

Function returns the predicted (fitted) values of the outcome variable using
the estimated posterior expectation of the linear covariate betas produced by
the <code>hdpGLM</code> function

Implementation of MCMC algorithms to estimate the Hierarchical Dirichlet Process Generalized Linear Model (hdpGLM) presented in the paper Ferrari (2020) Modeling Context-Dependent Latent Heterogeneity, Political Analysis <DOI:10.1017/pan.2019.13> and <doi:10.18637/jss.v107.i10>.

Diogo Ferrari

hdpGLM

Hierarchical Dirichlet Process Generalized Linear Models

predict.dpGLM function

<dl><dt>object</dt>
<dd>outcome of the function hdpLGM</dd>
<dt>new_data</dt>
<dd>data frame with the values of the covariates that are going
to be used to generate the predicted/fitted values. The posterior mean is
used to create the predicted values</dd>
<dt>...</dt>
<dd><code>family</code> : a string with the family of the output variable:
<code>gaussian</code> (default), <code>binomial</code>, etc...</dd></dl>

Arguments

dpGLM Predicted values — predict.dpGLM

<dl>

<dt>object</dt>
<dd>outcome of the function hdpLGM</dd>


<dt>new_data</dt>
<dd>data frame with the values of the covariates that are going
to be used to generate the predicted/fitted values. The posterior mean is
used to create the predicted values</dd>


<dt>...</dt>
<dd><code>family</code> : a string with the family of the output variable:
<code>gaussian</code> (default), <code>binomial</code>, etc...</dd>

</dl>

predict.dpGLM: dpGLM Predicted values

Description

Usage

Value

Arguments