buildAR: Builds an autoregressive model

In olshena/COVIDNearTerm: Near Term Prediction of COVID-Related Population Metrics

Builds an autoregressive model

Description

Description text

Usage

buildAR(
  vec,
  x = NULL,
  wsize,
  method = c("equal", "unweighted", "triangle"),
  seed = NULL,
  rhat_method = c("none", "geometric", "arithmetic"),
  lambda = 0
)

Arguments

vec	A vector of numeric data
x	A vector containing a covariate with which to generate AR model, if unspecified, defaults to vec
wsize	Number of prior observations to use for averaging
method	Type of weighting to use
seed	Seed for random number generator
rhat_method	Method for calculating rhat, if "none", rhat = 1 and has no effect
lambda	Shrinkage parameter, if not specified, default is 0 which produces no shrinkage. If an array is specified, all values in teh array are evaluated and the optimal lambda is chosen based on residual sum of squares. Values should be between 0 and 1 inclusive.

Value

A list with the following components:

wsize is the number of prior observations used for averaging

vec is a numeric vector containing the input data

fit is a numeric vector containing the autoregressive model fitted data

errors is a numeric vector containing the difference between fit and vec

initphi is a numeric vector containing the initial phi

finalphi is a numeric vector containing the final phi

olshena/COVIDNearTerm documentation built on May 27, 2023, 1:23 p.m.