ExampleDataBinary

A simulated data set containing 7994 observations, 5 continuous covariates and 100 providers.

datasets

Implements linear and generalized linear models for provider profiling,
incorporating both fixed and random effects. For large-scale providers, the linear
profiled-based method and the SerBIN method for binary data
reduce the computational burden. Provides post-modeling features, such as
indirect and direct standardization measures, hypothesis testing,
confidence intervals, and post-estimation visualization.
For more information, see Wu et al. (2022) <doi:10.1002/sim.9387>.

Xiaohan Liu

pprof

Modeling, Standardization and Testing for Provider Profiling

Lingfeng Luo

Yubo Shao

Xiangeng Fang

Wenbo Wu

Kevin He

ExampleDataBinary function

A list containing the following elements:<dl>
<dt>Y</dt>
<dd>a vector representing binary outcomes with 0 or 1.
Generated from a Bernoulli distribution with the probability of success (\(\boldsymbol{\mu}\)), which is
determined by applying the logistic function to the linear combination of provider effects and covariates.</dd><dt>ProvID</dt>
<dd>a vector representing identifier for each provider.
The number of individuals per provider is generated from a
Poisson distribution with mean 80, with a minimum value of 11.</dd><dt>Z</dt>
<dd>a data frame containing 5 continuous variables.
Generated from a multivariate normal distribution, where the mean is calculated as
\((\gamma_i - \mu_{\gamma}) \cdot \rho / \sigma_{\gamma}\) with \(\mu_{\gamma} = \log(4/11)\) and \(\sigma_{\gamma} = 0.4\),
and the correlation between covariates and provider effects is 0.1.</dd>
</dl>

Format

Example data with binary outcomes — ExampleDataBinary

A list containing the following elements:<dl>
<dt>Y</dt>
<dd>a vector representing binary outcomes with 0 or 1.
Generated from a Bernoulli distribution with the probability of success (\(\boldsymbol{\mu}\)), which is
determined by applying the logistic function to the linear combination of provider effects and covariates.</dd>

<dt>ProvID</dt>
<dd>a vector representing identifier for each provider.
The number of individuals per provider is generated from a
Poisson distribution with mean 80, with a minimum value of 11.</dd>

<dt>Z</dt>
<dd>a data frame containing 5 continuous variables.
Generated from a multivariate normal distribution, where the mean is calculated as
\((\gamma_i - \mu_{\gamma}) \cdot \rho / \sigma_{\gamma}\) with \(\mu_{\gamma} = \log(4/11)\) and \(\sigma_{\gamma} = 0.4\),
and the correlation between covariates and provider effects is 0.1.</dd>


</dl>

ExampleDataBinary: Example data with binary outcomes

Description

Usage

Arguments

Format

Examples