`calibration.Rd`

The implemented calibration tests for Poisson or negative binomial
predictions of count data are based on proper scoring rules and
described in detail in Wei and Held (2014).
The following proper scoring rules are available:
Dawid-Sebastiani score (`"dss"`

),
logarithmic score (`"logs"`

),
ranked probability score (`"rps"`

).

```
calibrationTest(x, ...)
# S3 method for default
calibrationTest(x, mu, size = NULL,
which = c("dss", "logs", "rps"),
tolerance = 1e-4, method = 2, ...)
```

- x
the observed counts. All involved functions are vectorized and also accept matrices or arrays.

- mu
the means of the predictive distributions for the observations

`x`

.- size
either

`NULL`

(default), indicating Poisson predictions with mean`mu`

, or dispersion parameters of negative binomial forecasts for the observations`x`

, parametrized as in`dnbinom`

with variance`mu*(1+mu/size)`

.- which
a character string indicating which proper scoring rule to apply.

- tolerance
absolute tolerance for the null expectation and variance of

`"logs"`

and`"rps"`

. For the latter, see the note below. Unused for`which = "dss"`

(closed form).- method
selection of the \(z\)-statistic:

`method = 2`

refers to the alternative test statistic \(Z_s^*\) of Wei and Held (2014, Discussion), which has been recommended for low counts.`method = 1`

corresponds to Equation 5 in Wei and Held (2014).- ...
unused (argument of the generic).

an object of class `"htest"`

,
which is a list with the following components:

a character string indicating the type of test
performed (including `which`

scoring rule).

a character string naming the supplied `x`

argument.

the \(z\)-statistic of the test.

the number of predictions underlying the test, i.e., `length(x)`

.

the p-value for the test.

If the gsl package is installed, its implementations of the
Bessel and hypergeometric functions are used when calculating the null
expectation and variance of the `rps`

.
These functions are faster and yield more accurate results (especially
for larger `mu`

).

Wei, W. and Held, L. (2014):
Calibration tests for count data.
*Test*, **23**, 787-805.

Sebastian Meyer and Wei Wei

```
mu <- c(0.1, 1, 3, 6, pi, 100)
size <- 0.1
set.seed(1)
y <- rnbinom(length(mu), mu = mu, size = size)
calibrationTest(y, mu = mu, size = size) # p = 0.99
calibrationTest(y, mu = mu, size = 1) # p = 4.3e-05
calibrationTest(y, mu = 1, size = size) # p = 0.6959
calibrationTest(y, mu = 1, size = size, which = "rps") # p = 0.1286
```