`extras`

provides helper functions for Bayesian analyses.

In particular it provides functions to summarise vectors of MCMC (Monte Carlo Markov Chain) samples, draw random samples from various distributions and calculate deviance residuals as well as R translations of some BUGS (Bayesian Using Gibbs Sampling), JAGS (Just Another Gibbs Sampler), STAN and TMB (Template Model Builder) functions.

## Installation

To install the developmental version from GitHub

```
# install.packages("pak")
pak::pak("poissonconsulting/extras")
```

## Demonstration

### Summarise MCMC Samples

The `extras`

package provides functions to summarise MCMC samples like `svalue()`

which gives the *surprisal value* (Greenland, 2019)

```
library(extras)
#>
#> Attaching package: 'extras'
#> The following object is masked from 'package:stats':
#>
#> step
set.seed(1)
x <- rnorm(100)
svalue(rnorm(100))
#> [1] 0.3183615
svalue(rnorm(100, mean = 1))
#> [1] 1.704015
svalue(rnorm(100, mean = 2))
#> [1] 3.850857
svalue(rnorm(100, mean = 3))
#> [1] 5.073249
```

### Distributions

Implemented distributions with functions to draw random samples, calculate log-likelihoods, and calculate deviance residuals for include:

- Bernoulli
- Beta-binomial
- Gamma
- Gamma-Poisson
- Zero-inflated gamma-Poisson
- Log-Normal
- Negative Binomial
- Normal
- Poisson
- Zero-inflated Poisson
- Skew Normal
- Student’s t

### R translations

The package also provides R translations of `BUGS`

(and `JAGS`

) functions such as `pow()`

and `log<-`

.

### Numericise R Objects

Atomic vectors, matrices, arrays and data.frames of appropriate classes can be converted to numeric objects suitable for Bayesian analysis using the `numericise()`

(and `numericize()`

) function.

```
numericise(
data.frame(
logical = c(TRUE, FALSE),
factor = factor(c("blue", "green")),
Date = as.Date(c("2000-01-01", "2000-01-02")),
hms = hms::as_hms(c("00:00:02", "00:01:01"))
)
)
#> logical factor Date hms
#> [1,] 1 1 10957 2
#> [2,] 0 2 10958 61
```

## References

Greenland, S. 2019. Valid P-Values Behave Exactly as They Should: Some Misleading Criticisms of P-Values and Their Resolution With S-Values. The American Statistician 73(sup1): 106–114.

## Code of Conduct

Please note that the extras project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.