ae_median_sample: Absolute error of the median (sample-based version)

In epiforecasts/scoringutils: Utilities for Scoring and Assessing Predictions

Absolute error of the median (sample-based version)

Description

Absolute error of the median calculated as

|\text{observed} - \text{median prediction}|

where the median prediction is calculated as the median of the predictive samples.

Usage

ae_median_sample(observed, predicted)

Arguments

observed	A vector with observed values of size n
predicted	nxN matrix of predictive samples, n (number of rows) being the number of data points and N (number of columns) the number of Monte Carlo samples. Alternatively, predicted can just be a vector of size n.

Value

Numeric vector of length n with the absolute errors of the median.

Input format

Overview of required input format for sample-based forecasts

See Also

ae_median_quantile()

Examples

observed <- rnorm(30, mean = 1:30)
predicted_values <- matrix(rnorm(30, mean = 1:30))
ae_median_sample(observed, predicted_values)

epiforecasts/scoringutils documentation built on Dec. 11, 2024, 11:12 a.m.