Arima_by_state: Use Rob Hyndman's forecast package to estimate drift in ARIMA...

In vjcitn/sars2app: COVID-19 data resources and analysis tools cut down for an interactive app

Description

Use Rob Hyndman's forecast package to estimate drift in ARIMA models

Usage

Arima_by_state(
  src,
  state.in = "New York",
  MAorder = NULL,
  Difforder = 1,
  basedate = "2020-02-15",
  lookback_days = 29,
  ARorder = NULL,
  max_date = NULL
)

Arguments

src	a tibble as returned by nytimes_state_data() or jhu_us_data()
state.in	character(1) state name
MAorder	numeric(1) order of moving average component
Difforder	numeric(1) order of differencing d in ARIMA(p,d,q)
basedate	character(1) used by lubridate::as_date to filter away all earlier records
lookback_days	numeric(1) only uses this many days from most recent in src
ARorder	order of autoregressive component
max_date	a date from which to start lookback ... defaults to NULL in which

Value

instance of S3 class Arima_sars2pack

Note

If ARIMA model fails, another attempt is made with lookback days halved. case the latest available date is used

Examples

nyd = nytimes_state_data()
mb = min_bic(nyd, state.in="New York")
lkny = Arima_by_state(nyd, ARorder=mb$opt["ARord"], MAorder=mb$opt["MAord"])
lkut = Arima_by_state(nyd, "Utah")
lkny
plot(lkny)
usd = jhu_us_data()
lkny2 = Arima_by_state(usd, ARorder=mb$opt["ARord"], MAorder=mb$opt["MAord"])
lkny2
plot(lkny2)
lknyNULL = Arima_by_state(nyd, ARorder=NULL, MAorder=NULL)
lknyNULL

vjcitn/sars2app documentation built on Jan. 3, 2022, 12:19 a.m.