scoring (version 0.6)

logscore: Calculate Logarithmic Scores

Description

Calculate logarithmic scores and average logarithmic scores by a grouping variable.

Usage

logscore(object, data, group = NULL, reverse = FALSE)

Arguments

object

an object of class "formula", of the form outcome ~ forecast. See calcscore() documentation for further details.

data

an optional data frame or list containing the variables in the formula. If not found in data, the variables are taken from the environment from which calcscore is called.

group

the name of a grouping variable within data, which is used to calculate average log score by group.

reverse

if FALSE (default), smaller scores imply better forecasts. If TRUE, larger scores imply better forecasts.

Value

Depending on input arguments, logscore may return an object of class numeric containing raw logarithmic scores. It may also return a list containing the objects below.

rawscores

an object of class numeric containing raw log scores for each forecast.

mnlog

an object of class numeric containing mean log scores for each unique value of group.

Details

If group is supplied, the function returns a list (see value section). Otherwise, the function returns a numeric vector containing the log score associated with each forecast.

The argument bounds is not available because the upper bound of the logarithmic score is infinity. If one wants a bounded rule that approximates the logarithmic rule, try using calcscore() with fam="pow" and param=1.001.

References

Toda, M. (1963). Measurement of subjective probability distributions. ESD-TDR-63-407, Decision Sciences Laboratory, L. G. Hanscom Field, Bedford, Mass.

Shuford, E. H., Albert, A., & Massengill, H. E. (1966). Admissible probability measurement procedures. Psychometrika, 31, 125-145.

See Also

calcscore

Examples

# NOT RUN {
data("WorldEvents")
## Raw log scores (0 best, infinity worst)
log1 <- logscore(answer ~ forecast, data=WorldEvents)
## Raw log scores (0 best, -infinity worst)
log1 <- logscore(answer ~ forecast, data=WorldEvents,
                 reverse = TRUE)
## Raw log scores plus group means
log2 <- logscore(answer ~ forecast, data=WorldEvents,
                 group="forecaster")
# }