Performance: R Package for Regression Model Assessment Metrics
by Daniel Lüdecke
Available on 1 platform
Sign in to view source links and access this dataset
Description
An R package providing utilities for computing model quality measures not directly available in R's base packages. It includes functions for metrics like R-squared, intraclass correlation coefficient, and root mean squared error, as well as checks for overdispersion and zero-inflation. The package, authored by Daniel Lüdecke, applies to a variety of regression models including generalized linear, mixed effects, and Bayesian models.
Use Cases
Calculate R-squared and intraclass correlation coefficients for regression models based on the provided functions.
Assess model fit and error using root mean squared error metrics described in the package.
Diagnose statistical issues like overdispersion or zero-inflation in generalized linear models.
Evaluate the performance of mixed effects or Bayesian regression models using the package's unified functions.
Strengths
Functions are based on cited statistical literature, including Nakagawa, Johnson & Schielzeth (2017).
Package is documented in a peer-reviewed journal article (Lüdecke et al., 2021, JOSS).
Supports a large variety of regression model types as stated in the description.
Limitations
License is closed, restricting commercial or open redistribution.
Column-level documentation and sample data are unavailable, requiring manual inspection after download.
Row count and dataset size are unknown, which may limit suitability assessment.
Provenance
Source
Daniel Lüdecke
Collection Method
Software package development for statistical computing.
Time Range
null
Freshness
Last update date is unknown; freshness unverified.