design: design

In AntoineDubois/sdcv2: RUM design Simulator

design

Description

The method of the Experiment class which generate a full factorial design

Arguments

name	The name of the experimental design to implement.
choice_set_size	The number of alternative per choice set.
clustered	Determines the way the data is represented in the experimental design table. 0 if the matrices of decision makers' characteristics and alternatives' attributes are row. 1 if the matrices of decision makers' characteristics and alternatives' attributes are clustered. 2 if the matrices of decision makers' characteristics and alternatives' attributes are categorized.
format	If "long" (default) the design is expressed in long format and wide format otherwise.

Value

an Experimental Design as well as a some pieces of information such as the D-score, defined as the determinant of the covariance matri<- of the preference parameter (a good D-score should be small), and an estimation of the preference parameters.

Examples

DM_att_names <- list("X1", "X2", "X3")
AT_att_names <- list("Z1", "Z2", "Z3")
AT_names <- list("good1", "good2", "good3", "good4")
groups <- c(10, 20)
FD <- Experiment(DM_att_names=DM_att_names, AT_att_names=AT_att_names, AT_names=AT_names, groups=groups)
FD$gen_AT_attributes()
FD$gen_DM_attributes()
FD$gen_preference_coefficients()
FD$utility()
FFD <- FD$design(choice_set_size=2, clustered=0) # generation of the full factorial design with row data
#by default, name="FuFD", choice_set_size = nb_alternatives
FFD1 <- FD$design(name="FuFD",choice_set_size=2, clustered=1, nb_levels_DM=c(3, 3, 4, 2), nb_levels_AT=c(3, 2, 2, 4), format="wide") # generation of the full factorial design with glustered data
FFD2 <- FD$design(choice_set_size=2, clustered=2, nb_levels_DM=c(2, 3, 4, 2), nb_levels_AT=c(2, 2, 2, 2)) # generation of the full factorial design with categorical data
FFD3 <- FD$design(name="FrFD", choice_set_size=2, clustered=2, nb_levels_DM=c(2, 3, 4, 2), nb_levels_AT=c(2, 2, 2, 2), nb_questions = 2) # Generation a a random fractional factorial design with categorical data
FFD4 <- FD$design(name="FrFD", choice_set_size=2, clustered=2, nb_levels_DM=c(2, 3, 4, 2), nb_levels_AT=c(3, 3, 3, 3), nb_questions = 2, format="wide") # Yet, we want to express this design in wide format

AntoineDubois/sdcv2 documentation built on May 16, 2024, 3:23 p.m.