designs_gen: GEnerate experimental design

In nikitagusarov/dcesimulatr: Discrete Choice Experiment Simulator

GEnerate experimental design

Description

Create experimental design using both factors and random specifications. The design configuration assumes that alternatives' attributes are mixed: vectors declared using 'c()' assignement and 'declared using 'r' class simulation functions.

Usage

designs_gen(experimental_design, n = NULL, resample = NULL)

Arguments

experimental_design	Input experimnetal design object
n	The n is not used in current version
resample	Declares whether the dataset should be resampled (shuffled) in case of factorial design element presence.

Value

data.frame A dataset of choice sets (Z) respecting FF Design.

Examples

# Create alternatives
alt1 <- alternative$new()
alt1$add_attributes(Quality = c(0:1), Price = rnorm(mean = 2, sd = 1))
alt2 <- alternative$new()
alt2$add_attributes(Size = c(1:3), Price = runif(min = 1, max = 3))

# Regroup alternatives into design
edesign <- experimental_design$new(alternatives = list(alt1, alt2), design = "mixed")
Z <- designs_gen(edesign, n = 2)

nikitagusarov/dcesimulatr documentation built on Jan. 7, 2023, 4:27 p.m.