sort_distributions {errint}

Sort Distributions by Better Error Intervals

Description

sort_distributions orders a given set of distribution assumptions in order of intervals accuracy error in ascending or descending order.

Usage

sort_distributions(phi, errors, dists = c("n", "nm", "l", "lm", "w", "b",
  "moge"), decreasing = FALSE, ...)

Arguments

phi

residual values used to compute the error interval.

errors

set of real errors corresponding to the predictions of a particular model.

dists

character vector with the distribution assumptions to test. See also 'Details'.

decreasing

logical, indicating whether or not distributions should be ordered by decreasing accuracy error.

...

additional arguments to be passed to functions error_interval and acc_intervals.

Details

Allowed distribution assumptions are:

• "n": Zero-mu Gaussian

• "nm": General Gaussian

• "l": Zero-mu Laplace

• "lm": General Laplace

• "b": Beta

• "w": Weibull

• "moge": Moge

Value

Returns an object of class c("df_intervals", "data.frame") with information of the distribution assumptions ordered by accuracy error.

Author(s)

Jesus Prada, [e-mail]

References

Link to the scientific paper

Prada, Jesus, and Jose Ramon Dorronsoro. "SVRs and Uncertainty Estimates in Wind Energy Prediction." Advances in Computational Intelligence. Springer International Publishing, 2015. 564-577,

with theoretical background for this package is provided below.

http://link.springer.com/chapter/10.1007/978-3-319-19222-2_47

See Also

df_intervals error_interval acc_intervals order

Examples

sort_distributions(rnorm(10),rnorm(10),decreasing=TRUE)