Entropy Function (Exported)

Extract entropy from a SurveyMixr object or compute from posterior probabilities.

Usage

entropy(object, weights = NULL)

# S3 method for class 'SurveyMixr'
entropy(object, weights = NULL)

# S3 method for class 'matrix'
entropy(object, weights = NULL)

# Default S3 method
entropy(object, weights = NULL)

Arguments

object

Either a SurveyMixr object or a matrix of posterior probabilities

weights

Optional survey weights for weighted entropy computation

Value

Numeric entropy value between 0 and 1, where values closer to 1 indicate better class separation

Details

Entropy measures the certainty of classification in mixture models. It is calculated as: $$E = 1 - \frac{\sum_i \sum_k p_{ik} \log(p_{ik})}{N \log(K)}$$

where \(p_{ik}\) is the posterior probability of individual i belonging to class k, N is the sample size, and K is the number of classes.

Values range from 0 (complete uncertainty, uniform classification) to 1 (perfect classification). Values above 0.80 generally indicate good class separation.

References

Ramaswamy, V., DeSarbo, W. S., Reibstein, D. J., & Robinson, W. T. (1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Marketing Science, 12(1), 103-124.

Examples

if (FALSE) { # \dontrun{
# From fitted model
entropy(fit)

# From posterior probability matrix
post_probs <- matrix(c(0.9, 0.1, 0.2, 0.8), nrow = 2, byrow = TRUE)
entropy(post_probs)
} # }