CorrStudySplit {SimEUCartelLaw}

Investigate the effect of correlated input parameters depending on illegal gain

Description

CorrStudySplit investigates the effect of correlated input parameters and its dependence on the illegal gain A.

Usage

CorrStudySplit(params, m = 1e+05, rho = seq(0.1, 0.9, by = 0.2),
  breaks = seq(0.1, 0.3, by = 0.04), QMC = FALSE, seed = 1)

Arguments

params

named list containing numeric vectors Phi, Rho, Chi, Ksi, M, G and A with the ranges for the input parameters.

m

numeric scalar containing the number of Monte Carlo replications (for each correlation intensity). Defaults to 1e5.

rho

a numeric vector containing correlation intensities. Defaults to seq(0.1,0.9,by=0.2).

breaks

a numeric vector with breaks for the construction of the intervals for the illegal gain A. Defaults to seq(0.1,0.3,by=0.04).

QMC

logical scalar. If TRUE, an equidistant grid is generated, if FALSE, uniformly distributed random numbers are simulated.

seed

numeric scalar containing the random seed for each simulation. Defaults to 1 in order to make results reproducible.

Details

CorrStudySplit performs repeated simulations via LEgame with different values for the correlation intensity and reports results for compliance and expected illegal gain for various subsets of simulated illegal gains A in order to further illustrate the effect of correlation on the deterrent effect of the legal exemption system.

Value

A matrix containing the results of the repeated simulations.

Examples

Par <- list(Phi=c(0.1,0.5), Rho=c(0.5,0.9), Ksi=c(0.05,0.3), Chi=c(0.1,0.4),
            M=c(0.2,1.2), G=c(0.05,0.2), A=c(0.1,0.3))
res <- CorrStudySplit(params=Par, m=10000)
print(res)