isodistrreg (version 0.1.0)

cdf: Cumulative distribution function (CDF) of IDR or raw forecasts

Description

Evaluate the the cumulative distribution function (CDF) of IDR predictions or of unprocessed forecasts in a data.frame.

Usage

cdf(predictions, thresholds)
# S3 method for idr
cdf(predictions, thresholds)
# S3 method for data.frame
cdf(predictions, thresholds)

Arguments

predictions

either an object of class idr (output of predict.idrfit), or a data.frame of numeric variables. In the latter case, the CDF is computed using the empirical distribution of the variables in predictions.

thresholds

numeric vector of thresholds at which the CDF will be evaluated.

Value

A matrix of probabilities giving the evaluated CDFs at the given thresholds, one column for each threshold.

Details

The CDFs are considered as piecewise constant stepfunctions: If x are the points where the IDR fitted CDF (or the empirical distribution of the forecasts) has jumps and p the corresponding CDF values, then for x[i] <= x < x[i + 1], the CDF at x is p[i].

See Also

predict.idrfit qpred, bscore

Examples

# NOT RUN {
data("rain")

## Postprocess HRES forecast using data of 3 years

X <- rain[1:(3 * 365), "HRES", drop = FALSE]
y <- rain[1:(3 * 365), "obs"]

fit <- idr(y = y, X = X)

## Compute probability of precipitation given that the HRES forecast is
## 0 mm, 0.5 mm or 1 mm

predictions <- predict(fit, data = data.frame(HRES = c(0, 0.5, 1)))
1 - cdf(predictions, thresholds = 0)
# }