cna (version 2.2.3)

makeFuzzy: Generate fuzzy-set data by simulating noise

Description

Generates fuzzy-set data by simulating the addition of random noise from the uncontrolled causal background to a data frame featuring binary factors only.

Usage

makeFuzzy(x, fuzzvalues = c(0, 0.05, 0.1), ...)

Arguments

x

Data frame or truthTab featuring binary factors with values 1 and 0 only.

fuzzvalues

Values to be added to the 0's and subtracted from the 1's.

…

Additional arguments are passed to truthTab.

Value

A truthTab of type "fs".

Details

In combination with allCombs, full.tt and selectCases, makeFuzzy is useful for simulating noisy data, which are needed for inverse search trials benchmarking the output of cna. makeFuzzy transforms a data frame or truthTab consisting of binary factors into a fuzzy-set truthTab by adding values selected at random from the argument fuzzvalues to the 0's and subtracting them from the 1's in the data frame. This transformation simulates the introduction of background noise into the data. selectCases can subsequently be applied to draw those fuzzy-set configurations from the resulting data that are compatible with a given data generating causal structure.

See Also

selectCases, allCombs, full.tt, truthTab, cna, tt2df, condition

Examples

# NOT RUN {
# Fuzzify a binary 6x3 matrix with default fuzzvalues.
X <- matrix(sample(0:1, 18, replace = TRUE), 6)
makeFuzzy(X)

# ... and with customized fuzzvalues.
makeFuzzy(X, fuzzvalues = 0:5/10)
makeFuzzy(X, fuzzvalues = seq(0, 0.45, 0.01))

# Generate all configurations of 5 fuzzy-set factors that are compatible with the causal
# structure A*b + C*D <-> E, such that con = .8 and cov = .8.
dat1 <- allCombs(c(2, 2, 2, 2, 2)) - 1
dat2 <- makeFuzzy(dat1, fuzzvalues = seq(0, 0.45, 0.01))
(dat3 <- selectCases1("A*b + C*D <-> E", con = .8, cov = .8, dat2))
condition("A*b + C*D <-> E", dat3)

# First, generate all configurations of 5 dichotomous factors that are compatible with
# the causal chain (A*b + a*B <-> C)*(C*d + c*D <-> E) and, second, introduce background
# noise. 
dat1 <- full.tt(5)
dat2 <- selectCases("(A*b + a*B <-> C)*(C*d + c*D <-> E)", dat1)
(dat3 <- makeFuzzy(dat2, fuzzvalues = seq(0, 0.45, 0.01)))
condition("(A*b + a*B <-> C)*(C*d + c*D <-> E)", dat3)

# Inverse search for the data generating causal structure A*b + a*B + C*D <-> E from
# fuzzy-set data with non-perfect consistency and coverage scores.
dat1 <- full.tt(5)
set.seed(55)
dat2 <- makeFuzzy(dat1, fuzzvalues = 0:4/10)
dat3 <- selectCases1("A*b + a*B + C*D <-> E", con = .8, cov = .8, dat2)
fscna(dat3, ordering = list("E"), strict = TRUE, con = .8, cov = .8)
# }