weight_of_evidence {island}

Weight of evidence

Description

weight_of_evidence calculates the weight of evidence of a set of nested models.

Usage

weight_of_evidence(data)

Arguments

data

A dataframe with the names of the models in the first column and their AIC values in the second column.

Details

Calculates the weight of evidence in favor of model i being the actual Kullback-Leibler best model given a set of models R for your data.

w_i = exp( - 1/2 * \Delta_i) / \Sigma exp( - 1/2 * \Delta_r)

r = 1, R

Value

A dataframe with the names of the analysed models, their AIC differences with respect to the best model and the w_i of each one.

References

K. P. Burnham, D. R. Anderson. Model selection and multimodel inference: a practical information-theoretic approach (New York:Springer, ed. 2, 2002).

See Also

akaikeic

Examples

models <- c("Best_3k", "Best_4k", "Best_5k", "Best_6k", "Best_7k",
  "Best_8k", "Best_9k")

  aks <- c(2977.852, 2968.568, 2957.384, 2952.618,
  2949.128, 2947.038, 2943.480)

  weight_of_evidence(cbind(models, aks))