1308979
doi
10.5281/zenodo.1308979
oai:zenodo.org:1308979
sjstats: Statistical Functions for Regression Models.
Daniel Lüdecke
Universitätsklinikum Hamburg-Eppendorf
url:https://github.com/strengejacke/sjstats/tree/0.16.0
info:eu-repo/semantics/openAccess
Other (Open)
<p>General</p>
<ul>
<li>The S3-generics for functions like <code>hdi()</code>, <code>rope()</code>, <code>equi_test()</code> etc. are now more generic, and function usage for each supported object is now included in the documentation.</li>
<li>Following functions are now S3-generic: <code>icc()</code>, <code>r2()</code>, <code>p_value()</code>, <code>se()</code>, and <code>std_beta()</code>.</li>
<li>Added <code>print()</code>-methods for some more functions, for a clearer output.</li>
<li>Revised <code>r2()</code> for mixed models (packages <strong>lme4</strong>, <strong>glmmTMB</strong>). The r-squared value should be much more precise now, and reports the marginal and conditional r-squared values.</li>
<li>Reduced package dependencies and removed <em>apaTables</em> and <em>MBESS</em> from suggested packages</li>
<li><code>stanmvreg</code>-models are now supported by many functions.</li>
</ul>
<p>New functions</p>
<ul>
<li><code>binned_resid()</code> to plot binned residuals for logistic regression models.</li>
<li><code>error_rate()</code> to compute model quality for logistic regression models.</li>
<li><code>auto_prior()</code> to quickly create automatically adjusted priors for brms-models.</li>
<li><code>difficulty()</code> to compute the item difficulty.</li>
</ul>
<p>Changes to functions</p>
<ul>
<li><code>icc()</code> gets a <code>ppd</code>-argument for Stan-models (<em>brmsfit</em> and <em>stanreg</em>), which performs a variance decomposition based on the posterior predictive distribution. This is the recommended way for non-Gaussian models.</li>
<li>For Stan-models (<em>brmsfit</em> and <em>stanreg</em>), <code>icc()</code> now also computes the HDI for the ICC and random-effect variances. Use the <code>prob</code>-argument to specify the limits of this interval.</li>
<li><code>link_inverse()</code> and <code>model_family()</code> now support <em>clmm</em>-models (package <em>ordinal</em>) and <em>glmRob</em> and <em>lmRob</em>-models (package <em>robust</em>).</li>
<li><code>model_family()</code> gets a <code>multi.resp</code>-argument, to return a list of family-informations for multivariate-response models (of class <code>brmsfit</code> or <code>stanmvreg</code>).</li>
<li><code>link_inverse()</code> gets a <code>multi.resp</code>-argument, to return a list of link-inverse-functions for multivariate-response models (of class <code>brmsfit</code> or <code>stanmvreg</code>).</li>
<li><code>p_value()</code> now supports <em>rlm</em>-models (package <em>MASS</em>).</li>
<li><code>check_assumptions()</code> for single models with <code>as.logical = FALSE</code> now has a nice print-method.</li>
<li><code>eta_sq()</code> and <code>omega_sq()</code> now also work for repeated-measure Anovas, i.e. Anova with error term (requires broom > 0.4.5).</li>
</ul>
<p>Bug fixes</p>
<ul>
<li><code>model_frame()</code> and <code>var_names()</code> now correctly cleans nested patterns like <code>offset(log(x + 10))</code> from column names.</li>
<li><code>model_frame()</code> now returns proper column names from <em>gamm4</em> models.</li>
<li><code>model_frame()</code> did not work when the model frame had spline-terms and weights.</li>
<li>Fix issue in <code>robust()</code> when <code>exponentiate = TRUE</code> and <code>conf.int = FALSE</code>.</li>
<li><code>reliab_test()</code> returned an error when the provided data frame has less than three columns, instead of returning <code>NULL</code>.</li>
</ul>
Zenodo
2018-07-10
info:eu-repo/semantics/other
1284472
0.16.0
1579937237.889607
778679
md5:6d0ba59bcb5b01c5b105b4dd3d2cb300
https://zenodo.org/records/1308979/files/strengejacke/sjstats-0.16.0.zip
public
https://github.com/strengejacke/sjstats/tree/0.16.0
Is supplement to
url
10.5281/zenodo.1284472
isVersionOf
doi