confint

confint.SMN

confint.SMSN

Computes either asymptotical (based on normality from maximum likelihood estimators) or parametric bootstraped confidence intervals from a model fit.

models

It fits scale mixture of skew-normal linear mixed models using either an expectation–maximization (EM) type algorithm or its accelerated version (Damped Anderson Acceleration with Epsilon Monotonicity, DAAREM), including some possibilities for modeling the within-subject dependence <doi:10.18637/jss.v115.i07>.

Fernanda L Schumacher

skewlmm

Scale Mixture of Skew-Normal Linear Mixed Models

Fernanda L. Schumacher

Larissa A. Matos

Victor H. Lachos

Katherine A. L. Valeriano

Nicholas Henderson

Ravi Varadhan

confint function

<dl> <dt>object</dt>
<dd>An object inheriting from class <code>SMN</code> or <code>SMSN</code></dd> <dt>parm</dt>
<dd>A character indicating for which parameter the intervals should be returned. Available options: "beta" for fixed effects, or "all" for all parameters. Default is "all".</dd> <dt>level</dt>
<dd>Confidence level to be used.</dd> <dt>method</dt>
<dd>A character indicating with method should be used. "asymptotic" refers to traditional confidence intervals based on asymptotical normality from maximum likelihood estimators; "bootstrap" performs a parametric bootstrap method based on B samples (100 by default), and is only recommended to small to moderate sample sizes.</dd> <dt>...</dt>
<dd>Additional arguments to be passed to <code>boot_par</code>.</dd> </dl>

Arguments

Computes confidence intervals from smn.lmm and smsn.lmm fitted models
 — confint

<dl>

 <dt>object</dt>
<dd>An object inheriting from class <code>SMN</code> or <code>SMSN</code></dd>

 <dt>parm</dt>
<dd>A character indicating for which parameter the intervals should be returned. Available options: "beta" for fixed effects, or "all" for all parameters. Default is "all".</dd>

 <dt>level</dt>
<dd>Confidence level to be used.</dd>

 <dt>method</dt>
<dd>A character indicating with method should be used. "asymptotic" refers to traditional confidence intervals based on asymptotical normality from maximum likelihood estimators; "bootstrap" performs a parametric bootstrap method based on B samples (100 by default), and is only recommended to small to moderate sample sizes.</dd>

 <dt>...</dt>
<dd>Additional arguments to be passed to <code>boot_par</code>.</dd>

 </dl>

Computes confidence intervals from smn.lmm and smsn.lmm fitted models

confint: Computes confidence intervals from smn.lmm and smsn.lmm fitted models

Description

Usage

Value

Arguments

See Also

Examples