MI2: Calculates mutual information

In nvihrs14/tcherry: Learning the structure of tcherry trees

Description

Calculate mutual information for two or three categorical variables.

Usage

MI2(x, y, smooth = 0, log_base = 2)

MI3(x, y, z, smooth = 0, log_base = 2)

Arguments

x, y, z	Vectors of class character or factor.
smooth	Additional cell counts for bayesian estimation of probabilities.
log_base	The base of the logarithmic function to be used.

Details

The mutual information for two variables is calculated by the formula

MI(x, y) = ∑ P(x, y) log(P(x, y) / (P(x)P(y)))

where the sum is over alle possible values of x and y.

The mutual information for three variables is calculated by the formula

MI(x, y, z) = ∑ P(x, y, z) log(P(x, y, z) / (P (x)P(y)P(z)))

where the sum is over all possible values of x, y and z.

Value

The mutual information given by a single numeric value.

Author(s)

Katrine Kirkeby, [e-mail]

Maria Knudsen, [e-mail]

Ninna Vihrs, [e-mail]

References

\insertRef

TCJTtcherry \insertRefEKTStcherry

See Also

MIk for mutual information for k variables.

Examples

var1 <- c(sample(c(1, 2), 100, replace = TRUE))
var2 <- var1 + c(sample(c(1, 2), 100, replace = TRUE))
var3 <- c(sample(c(1, 2), 100, replace = TRUE))
var1 <- as.character(var1)
var2 <- as.character(var2)
var3 <- as.character(var3)

MI2(var1, var2, smooth = 1)
MI2(var1, var2, smooth = 0.1, log_base = exp(1))

MI3(var1, var2, var3, smooth = 1)
MI3(var1, var2, var3, smooth = 0.1, log_base = exp(1))

nvihrs14/tcherry documentation built on Aug. 1, 2020, 6:25 p.m.