SHELF (version 1.2.0)

elicitConcProb: Elicit a concordance probability for two uncertain quantities, and plot a joint sample

Description

Given two elicited marginal distributions, open a browser in which one specifies a quadrant probability P(X_1 > m_1, X_2 > m_2), where m_1 and m_2 are the elicited medians of X_1 and X_2. A joint sample from the distribution of X_1 and X_2 is generated, using the two elicited marginal distributions and a bivariate normal copula.

Usage

elicitConcProb(fit1, fit2, m1, m2, d = c("best", "best"), n = 10000)

Arguments

fit1

An elicitation fit produced from the fitdist command for the first uncertain quantity X_1.

fit2

An elicitation fit produced from the fitdist command for the second uncertain quantity X_2.

m1

The elicited (or fitted) median of X_1.

m2

The elicited (or fitted) median of X_2.

d

A vector of distributions to be used for each elicited quantity: a string with elements chosen from "Normal", "Student-t", "Gamma", "Log normal" "Log Student-t", "Beta". The default is to use the best fitting distribution in each case.

n

The number of sampled (X_1, X_2) pairs to be plotted.

Value

A matrix of sampled values, one row per sample.

Examples

## Not run: 
# p1 <- c(0.25, 0.5, 0.75)
# v1 <- c(0.5, 0.55, 0.6)
# v2 <- c(0.22, 0.3, 0.35)
# myfit1 <- fitdist(v1, p1, 0, 1)
# myfit2 <- fitdist(v2, p1, 0, 1)
# elicitQuadProb(myfit1, myfit2, 0.55, 0.3, d=c("Beta", "Beta"))
# ## End(Not run)