SortQuantiles: Sort multiple quantiles to remove quantile crossing

In jbrowell/ProbCast: Flexible Probabilistic Forecasting

Sort multiple quantiles to remove quantile crossing

Description

This function sorts quantiles so that q1<q2<q3<...

Usage

SortQuantiles(data, Limits = NULL)

Arguments

data	A MultiQR containing multiple quantiles
Limits	A list of the upper (U) and lower (L) limits to apply to quantiles if the output should be bounded. E.g. list(U=0.999,L=0.001)/

Details

The method use in the production of multiple quantiles may not guarantee that quantiles are (non-strictly) increasing, which is necessary if combining quantiles to form a cumulative distribution function. Similarly, the quantiles may not respect the boundaries of the target variable. This function re-orders quantiles and constrains them to the user-specified limits.

Value

A MultiQR, data.frame object with (non-strictly) increasing quanitles.

Author(s)

Jethro Browell, [e-mail]

jbrowell/ProbCast documentation built on July 20, 2024, 1:53 p.m.