rqssoptimize: Optimize llambda (AQR) and alpha (LOESS) smoothing parameters...

In jameswcraig/binlessVPC: A Package to Implement Binless VPC using Regression Techniques

Description

Optimize llambda (AQR) and alpha (LOESS) smoothing parameters for observed data.

Usage

rqssoptimize(o, quant = c(0.05, 0.5, 0.95), interval = c(0, 7),
  pred.correct = FALSE, ...)

Arguments

o	A vpcstatsobj which includes both observed and simulated data.
quant	A numeric vector of length three containing prediction intervals.
interval	Interval range for lambda optimization.
pred.correct	Set TRUE to return pcVPC.
...	Other arguments to include

Value

Returns a data.table in vpcstatsobj containing optimized lambda smoothing parameters for each quantile.

Examples

library(data.table)
library(quantreg)
library(magrittr)

vpc <- observed(obswarfec, x=TIME, y=DV) %>%
simulated(simwarfec, y=DV) %>%
rqssoptimize(quant = c(0.1, 0.5, 0.9)) %>%
rqsssummarize() %>%
binlessplot()

pcvpc <- observed(obspheno, x=TIME, y=DV, pred = PRED) %>%
simulated(simpheno, y=DV, pred = PRED) %>%
rqssoptimize(quant = c(0.1, 0.50, 0.9), pred.correct = TRUE) %>%
rqsssummarize() %>%
binlessplot()

jameswcraig/binlessVPC documentation built on Nov. 10, 2019, 10:32 p.m.