performance

performance: An R Package for Assessment, Comparison and Testing of Statistical Models - Published in JOSS (2021)

https://github.com/easystats/performance

Keywords

Keywords from Contributors

Scientific Fields

Repository

:muscle: Models' quality and performance metrics (R2, ICC, LOO, AIC, BF, ...)

README.Rmd

---
output: github_document
bibliography: paper.bib
---

# performance 

```{r, echo = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  warning = FALSE,
  message = FALSE,
  out.width = "100%",
  dpi = 150,
  fig.path = "man/figures/",
  comment = "#>"
)

options(
  knitr.kable.NA = "",
  digits = 4,
  width = 100
)

library(performance)
```

[![DOI](https://joss.theoj.org/papers/10.21105/joss.03139/status.svg)](https://doi.org/10.21105/joss.03139)
[![downloads](https://cranlogs.r-pkg.org/badges/performance)](https://cran.r-project.org/package=performance) [![total](https://cranlogs.r-pkg.org/badges/grand-total/performance)](https://cranlogs.r-pkg.org/)

***Test if your model is a good model!***

A crucial aspect when building regression models is to evaluate the quality of modelfit. It is important to investigate how well models fit to the data and which fit indices to report. Functions to create diagnostic plots or to compute fit measures do exist, however, mostly spread over different packages. There is no unique and consistent approach to assess the model quality for different kind of models.

The primary goal of the **performance** package is to fill this gap and to provide utilities for computing **indices of model quality** and **goodness of fit**. These include measures like r-squared (R2), root mean squared error (RMSE) or intraclass correlation coefficient (ICC) , but also functions to check (mixed) models for overdispersion, zero-inflation, convergence or singularity.

## Installation

[![CRAN](https://www.r-pkg.org/badges/version/performance)](https://cran.r-project.org/package=performance) [![performance status badge](https://easystats.r-universe.dev/badges/performance)](https://easystats.r-universe.dev) [![codecov](https://codecov.io/gh/easystats/performance/branch/main/graph/badge.svg)](https://app.codecov.io/gh/easystats/performance)

The *performance* package is available on CRAN, while its latest development version is available on R-universe (from _rOpenSci_).

Type | Source | Command
---|---|---
Release | CRAN | `install.packages("performance")`
Development | R-universe | `install.packages("performance", repos = "https://easystats.r-universe.dev")`

Once you have downloaded the package, you can then load it using:

```{r, eval=FALSE}
library("performance")
```

> **Tip**
>
> Instead of `library(performance)`, use `library(easystats)`. This will make all features of the easystats-ecosystem available.
>
> To stay updated, use `easystats::install_latest()`.

## Citation

To cite performance in publications use:

```{r}
citation("performance")
```

## Documentation

[![Documentation](https://img.shields.io/badge/documentation-performance-orange.svg?colorB=E91E63)](https://easystats.github.io/performance/)
[![Blog](https://img.shields.io/badge/blog-easystats-orange.svg?colorB=FF9800)](https://easystats.github.io/blog/posts/)
[![Features](https://img.shields.io/badge/features-performance-orange.svg?colorB=2196F3)](https://easystats.github.io/performance/reference/index.html)


There is a nice introduction into the package on [youtube](https://www.youtube.com/watch?v=EPIxQ5i5oxs).

## The *performance* workflow

```{r workflow, echo=FALSE, out.width="75%"}
knitr::include_graphics("man/figures/figure_workflow.png")
```

### Assessing model quality

#### R-squared

**performance** has a generic `r2()` function, which computes the r-squared for
many different models, including mixed effects and Bayesian regression models.

`r2()` returns a list containing values related to the "most appropriate"
r-squared for the given model.

```{r}
model <- lm(mpg ~ wt + cyl, data = mtcars)
r2(model)

model <- glm(am ~ wt + cyl, data = mtcars, family = binomial)
r2(model)

library(MASS)
data(housing)
model <- polr(Sat ~ Infl + Type + Cont, weights = Freq, data = housing)
r2(model)
```

The different R-squared measures can also be accessed directly via functions like `r2_bayes()`, `r2_coxsnell()` or `r2_nagelkerke()` (see a full list of functions [here](https://easystats.github.io/performance/reference/index.html#section-r-functions)).

For mixed models, the _conditional_ and _marginal_ R-squared are returned. The
_marginal R-squared_ considers only the variance of the fixed effects and
indicates how much of the model's variance is explained by the fixed effects
part only. The _conditional R-squared_ takes both the fixed and random effects
into account and indicates how much of the model's variance is explained by the
"complete" model.

For frequentist mixed models, `r2()` (resp. `r2_nakagawa()`) computes the _mean_
random effect variances, thus `r2()` is also appropriate for mixed models with
more complex random effects structures, like random slopes or nested random
effects [@johnson_extension_2014; @nakagawa_coefficient_2017].

```{r}
set.seed(123)
library(rstanarm)

model <- stan_glmer(
  Petal.Length ~ Petal.Width + (1 | Species),
  data = iris,
  cores = 4
)

r2(model)

library(lme4)
model <- lmer(Reaction ~ Days + (1 + Days | Subject), data = sleepstudy)
r2(model)
```

#### Intraclass Correlation Coefficient (ICC)

Similar to R-squared, the ICC provides information on the explained variance and
can be interpreted as "the proportion of the variance explained by the grouping
structure in the population" [@hox_multilevel_2010].

`icc()` calculates the ICC for various mixed model objects, including `stanreg`
models.

```{r}
library(lme4)
model <- lmer(Reaction ~ Days + (1 + Days | Subject), data = sleepstudy)
icc(model)
```

...and models of class `brmsfit`.

```{r, echo=FALSE, eval=curl::has_internet()}
model <- insight::download_model("brms_mixed_1")
```
```{r, eval=FALSE}
library(brms)
set.seed(123)
model <- brm(mpg ~ wt + (1 | cyl) + (1 + wt | gear), data = mtcars)
```
```{r}
icc(model)
```

### Model diagnostics

#### Check for overdispersion

Overdispersion occurs when the observed variance in the data is higher than the
expected variance from the model assumption (for Poisson, variance roughly
equals the mean of an outcome). `check_overdispersion()` checks if a count model
(including mixed models) is overdispersed or not.

```{r}
library(glmmTMB)
data(Salamanders)
model <- glm(count ~ spp + mined, family = poisson, data = Salamanders)
check_overdispersion(model)
```

Overdispersion can be fixed by either modelling the dispersion parameter (not
possible with all packages), or by choosing a different distributional family
(like Quasi-Poisson, or negative binomial, see [@gelman_data_2007]).

#### Check for zero-inflation

Zero-inflation (in (Quasi-)Poisson models) is indicated when the amount of
observed zeros is larger than the amount of predicted zeros, so the model is
_underfitting_ zeros. In such cases, it is recommended to use negative binomial
or zero-inflated models.

Use `check_zeroinflation()` to check if zero-inflation is present in the fitted model.

```{r}
model <- glm(count ~ spp + mined, family = poisson, data = Salamanders)
check_zeroinflation(model)
```

#### Check for singular model fits

A "singular" model fit means that some dimensions of the variance-covariance
matrix have been estimated as exactly zero. This often occurs for mixed models
with overly complex random effects structures.

`check_singularity()` checks mixed models (of class `lme`, `merMod`, `glmmTMB`
or `MixMod`) for singularity, and returns `TRUE` if the model fit is singular.

```{r}
library(lme4)
data(sleepstudy)

# prepare data
set.seed(123)
sleepstudy$mygrp <- sample(1:5, size = 180, replace = TRUE)
sleepstudy$mysubgrp <- NA
for (i in 1:5) {
  filter_group <- sleepstudy$mygrp == i
  sleepstudy$mysubgrp[filter_group] <-
    sample(1:30, size = sum(filter_group), replace = TRUE)
}

# fit strange model
model <- lmer(
  Reaction ~ Days + (1 | mygrp / mysubgrp) + (1 | Subject),
  data = sleepstudy
)

check_singularity(model)
```

Remedies to cure issues with singular fits can be found [here](https://easystats.github.io/performance/reference/check_singularity.html).

#### Check for heteroskedasticity

Linear models assume constant error variance (homoskedasticity).

The `check_heteroscedasticity()` functions assess if this assumption has been
violated:

```{r}
data(cars)
model <- lm(dist ~ speed, data = cars)

check_heteroscedasticity(model)
```

#### Comprehensive visualization of model checks

**performance** provides many functions to check model assumptions, like
`check_collinearity()`, `check_normality()` or `check_heteroscedasticity()`. To
get a comprehensive check, use `check_model()`.

```{r, fig.height=12, fig.width=10, out.width="80%"}
# defining a model
model <- lm(mpg ~ wt + am + gear + vs * cyl, data = mtcars)

# checking model assumptions
check_model(model)
```

### Model performance summaries

`model_performance()` computes indices of model performance for regression
models. Depending on the model object, typical indices might be r-squared, AIC,
BIC, RMSE, ICC or LOOIC.

#### Linear model

```{r}
m1 <- lm(mpg ~ wt + cyl, data = mtcars)
model_performance(m1)
```

#### Logistic regression

```{r}
m2 <- glm(vs ~ wt + mpg, data = mtcars, family = "binomial")
model_performance(m2)
```

#### Linear mixed model

```{r}
library(lme4)
m3 <- lmer(Reaction ~ Days + (1 + Days | Subject), data = sleepstudy)
model_performance(m3)
```

### Models comparison

The `compare_performance()` function can be used to compare the performance and
quality of several models (including models of different types).

```{r}
counts <- c(18, 17, 15, 20, 10, 20, 25, 13, 12)
outcome <- gl(3, 1, 9)
treatment <- gl(3, 3)
m4 <- glm(counts ~ outcome + treatment, family = poisson())

compare_performance(m1, m2, m3, m4, verbose = FALSE)
```

#### General index of model performance

One can also easily compute and a [**composite index**](https://easystats.github.io/performance/reference/compare_performance.html#details) of model performance and sort the models from the best one to the worse.

```{r}
compare_performance(m1, m2, m3, m4, rank = TRUE, verbose = FALSE)
```

#### Visualisation of indices of models' performance

Finally, we provide convenient visualisation (the `see` package must be
installed).

```{r}
plot(compare_performance(m1, m2, m4, rank = TRUE, verbose = FALSE))
```

### Testing models

`test_performance()` (and `test_bf`, its Bayesian sister) carries out the most
relevant and appropriate tests based on the input (for instance, whether the
models are nested or not).

```{r}
set.seed(123)
data(iris)

lm1 <- lm(Sepal.Length ~ Species, data = iris)
lm2 <- lm(Sepal.Length ~ Species + Petal.Length, data = iris)
lm3 <- lm(Sepal.Length ~ Species * Sepal.Width, data = iris)
lm4 <- lm(Sepal.Length ~ Species * Sepal.Width + Petal.Length + Petal.Width, data = iris)

test_performance(lm1, lm2, lm3, lm4)

test_bf(lm1, lm2, lm3, lm4)
```

### Plotting Functions

Plotting functions are available through the [**see** package](https://easystats.github.io/see/articles/performance.html).


# Code of Conduct

Please note that the performance project is released with a [Contributor Code of Conduct](https://easystats.github.io/performance/CODE_OF_CONDUCT.html). By contributing to this project, you agree to abide by its terms.


# Contributing

We are happy to receive bug reports, suggestions, questions, and (most of all)
contributions to fix problems and add features.

Please follow contributing guidelines mentioned here:



## References

Owner

JOSS Publication

performance: An R Package for Assessment, Comparison and Testing of Statistical Models

Published

April 21, 2021

DOI

10.21105/joss.03139

Volume 6, Issue 60, Page 3139

Authors

Daniel Lüdecke
University Medical Center Hamburg-Eppendorf, Germany

Mattan S. Ben-Shachar
Ben-Gurion University of the Negev, Israel

Indrajeet Patil
Center for Humans and Machines, Max Planck Institute for Human Development, Berlin, Germany

Philip Waggoner
University of Chicago, USA

Dominique Makowski
Nanyang Technological University, Singapore

Editor

Mikkel Meyer Andersen

Tags

easystats parameters regression linear models coefficients

View PDF Review Thread Software Archive

Papers & Mentions

GitHub Events

Committers

Last synced: 8 months ago

Issues and Pull Requests

Last synced: 4 months ago

Packages