Title:	Utilities for Aggregating Probabilistic Forecasts
Version:	1.0.2
URL:	https://github.com/forecastingresearch/aggutils
BugReports:	https://github.com/forecastingresearch/aggutils/issues
Description:	Provides several methods for aggregating probabilistic forecasts. You have a group of people who have made probabilistic forecasts for the same event. You want to take advantage of the "wisdom of the crowd" and combine these forecasts in some sensible way. This package provides implementations of several strategies, including geometric mean of odds, an extremized aggregate (Neyman, Roughgarden (2021) <doi:10.1145/3490486.3538243>), and "high-density trimmed mean" (Powell et al. (2022) <doi:10.1037/dec0000191>).
License:	MIT + file LICENSE
Encoding:	UTF-8
RoxygenNote:	7.2.3
Imports:	stats, docstring
Suggests:	testthat
NeedsCompilation:	no
Packaged:	2023-08-22 14:54:52 UTC; molly
Author:	Molly Hickman [aut, cre], Zach Jacobs [aut]
Maintainer:	Molly Hickman <molly@forecastingresearch.org>
Repository:	CRAN
Date/Publication:	2023-08-22 18:30:12 UTC

Geometric Mean

Description

Calculate the geometric mean of a vector of forecasts. We handle 0s by replacing them with the qth quantile of the non-zero forecasts.

Usage

geoMeanCalc(x, q = 0.05)

Arguments

Value

(numeric) The geometric mean of the vector

Note

agg(a) + agg(not a) does not sum to 1 for this aggregation method.

Geometric Mean of Odds

Description

Convert probabilities to odds, and calculate the geometric mean of the odds. We handle 0s by replacing them with the qth quantile of the non-zero forecasts, before converting.

Usage

geoMeanOfOddsCalc(x, q = 0.05, odds = FALSE)

Arguments

Value

(numeric) The geometric mean of the odds

Note

agg(a) + agg(not a) does not sum to 1 for this aggregation method.

Highest-Density Trimmed Mean

Description

From Powell et al. (2022) doi:10.1037/dec0000191. You find the shortest interval containing (1-p) * 100% of the data and take the mean of the forecasts within that interval.

Usage

hd_trim(x, p = 0.1)

Arguments

Value

(numeric) The highest-density trimmed mean of the vector

Note

As p gets bigger this acts like a mode in a similar way to the symmetrically-trimmed mean acting like a median.

Neyman Aggregation (Extremized)

Description

Takes the arithmetic mean of the log odds of the forecasts, then extremizes the mean by a factor d, where d is

(n*(sqrt((3n^2) - (3n) + 1) - 2))/(n^2 - n - 1)

where n is the number of forecasts.

Usage

neymanAggCalc(x)

Arguments

Value

(numeric) The extremized mean of the vector

References

Neyman, E. and Roughgarden, T. (2021). Are you smarter than a random expert? The robust aggregation of substitutable signals: doi:10.1145/3490486.3538243. Also Jaime Sevilla's EAF post “Principled extremizing of aggregated forecasts."

Preprocessing function for agg methods

Description

This does the preprocessing steps that all the agg methods have in common.

Usage

preprocess(x, q = 0)

Arguments

Value

A vector of forecasts with 0s are replaced by the qth quantile and 100s are replaced by the (1 - q)th quantile.

Note

Assumes forecasts are in the range 0 to 100, inclusive.

Soften the mean.

Description

If the mean is > .5, trim the top trim%; if < .5, the bottom trim%. Return the new mean (i.e. soften the mean).

Usage

soften_mean(x, p = 0.1)

Arguments

Value

(numeric) The softened mean of the vector

Note

This goes against usual wisdom of extremizing the mean, but performs well when the crowd has some overconfident forecasters in it.

Trimmed mean

Description

Trim the top and bottom (p*100)% of forecasts

Usage

trim(x, p = 0.1)

Arguments

Value

(numeric) The trimmed mean of the vector