SHELF-package: Tools to Support the Sheffield Elicitation Framework

In OakleyJ/SHELF: Tools to Support the Sheffield Elicitation Framework

Tools to Support the Sheffield Elicitation Framework

Description

Implements various methods for eliciting a probability distribution for a single parameter from an expert or a group of experts. The expert provides a small number of probability judgements, corresponding to points on his or her cumulative distribution function. A range of parametric distributions can then be fitted and displayed, with feedback provided in the form of fitted probabilities and percentiles. For multiple experts, a weighted linear pool can be calculated. Also includes functions for eliciting beliefs about population distributions, eliciting multivariate distributions using a Gaussian copula, eliciting a Dirichlet distribution, and eliciting distributions for variance parameters in a random effects meta-analysis model. R Shiny apps for most of the methods are included.

Package:	SHELF
Type:	Package
Version:	1.9.0.9000
Date:	2023-05-31
License:	GPL-2 | GPL-3

Author(s)

Jeremy Oakley <[e-mail]>

References

The SHELF homepage

Examples

## Not run: 
## 1) Elicit judgements from two experts individually 
# Expert A states P(X<30)=0.25, P(X<40)=0.5, P(X<50)=0.75
# Expert B states P(X<20)=0.25, P(X<25)=0.5, P(X<35)=0.75
# Both experts state 0<X<100.

## 2) Fit distributions to each expert's judgements
v <- matrix(c(30, 40, 50, 20, 25, 35), 3, 2)
p <- c(0.25, 0.5, 0.75)
myfit <- fitdist(vals = v, probs = p, lower = 0, upper = 100)

## 3) Plot the fitted distributions, including a linear pool
plotfit(myfit, lp = T)

## 4) Now elicit a single 'consensus' distribution from the two experts
# Suppose they agree P(X<25)=0.25, P(X<30)=0.5, P(X<40)=0.75
v <-c(25, 30, 40)
p <-c(0.25, 0.5, 0.75)
myfit <- fitdist(vals = v, probs = p, lower = 0, upper = 100)

## 5) Plot the fitted density, and report some feedback, such as the 
# fitted 5th and 95th percentiles
plotfit(myfit, ql = 0.05, qu = 0.95)
feedback(myfit, quantiles = c(0.05, 0.95))

## Can also use interactive plotting
v <- matrix(c(30, 40, 50, 20, 25, 35), 3, 2)
p <- c(0.25, 0.5, 0.75)
myfit <- fitdist(vals = v, probs = p, lower = 0, upper = 100)
# plot each distribution
plotfit(myfit)

## plot the distribution for one expert only
plotfit(myfit, ex = 1)

## Enter judgements in interactive mode
elicit()

#' ## Enter separate judgements for each expert in interactive mode
elicitMultiple()


## End(Not run)

OakleyJ/SHELF documentation built on March 17, 2024, 8:13 p.m.