log_shift: Log transformation with an additive shift

In epiforecasts/scoringutils: Utilities for Scoring and Assessing Predictions

Log transformation with an additive shift

Description

Function that shifts a value by some offset and then applies the natural logarithm to it.

Usage

log_shift(x, offset = 0, base = exp(1))

Arguments

x	vector of input values to be transformed
offset	Number to add to the input value before taking the natural logarithm.
base	A positive number: the base with respect to which logarithms are computed. Defaults to e = exp(1).

Details

The output is computed as log(x + offset)

Value

A numeric vector with transformed values

References

Transformation of forecasts for evaluating predictive performance in an epidemiological context Nikos I. Bosse, Sam Abbott, Anne Cori, Edwin van Leeuwen, Johannes Bracher, Sebastian Funk medRxiv 2023.01.23.23284722 \Sexpr[results=rd]{tools:::Rd_expr_doi("https://doi.org/10.1101/2023.01.23.23284722")} https://www.medrxiv.org/content/10.1101/2023.01.23.23284722v1 # nolint

Examples

log_shift(1:10)
log_shift(0:9, offset = 1)

transform_forecasts(
  as_forecast(example_quantile)[observed > 0, ],
  fun = log_shift,
  offset = 1
 )

epiforecasts/scoringutils documentation built on April 23, 2024, 4:56 p.m.