PPD_2_MultiQR: Produce a MultiQR object from a Parametric Predictive...

In jbrowell/ProbCast: Flexible Probabilistic Forecasting

Produce a MultiQR object from a Parametric Predictive Distribution

Description

Produce multiple predictive quantiles as a MultiQR object from a PPD object. Alternatively return predicted parameters from the PPD models.

Note that this function may be superseded by an S3 method in future versions of ProbCast.

Usage

PPD_2_MultiQR(data, models, quantiles = seq(0.05, 0.95, by = 0.05), params = F)

Arguments

data	A data.frame containing explanatory variables required by models.
models	A PPD object.
quantiles	Vector of quantiles to be included in the returned MultiQR object.

Details

Exact quantiles of the (semi-) parametric models (predictive distributions) are calculated and output as a MultiQR object. Warnings are thrown if the input data result in NA estimates of quantiles, likely as a result of inputs being outside the allowable range for the PPD model.

Value

A MultiQR object derived from gamlss predictive distributions. Alternatively, a matrix containing the parameters of the predictive gamlss distributions.

Author(s)

Jethro Browell, [e-mail]

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