stsm_filter {autostsm}

Kalman Filter

Description

Kalman filter an estimated model from stsm_estimate output. This is a wrapper to stsm_forecast with n.ahead = 0.

Usage

stsm_filter(
  model,
  y,
  freq = NULL,
  exo_obs = NULL,
  exo_state = NULL,
  ci = 0.8,
  plot = FALSE,
  plot.decomp = FALSE,
  n.hist = NULL,
  smooth = TRUE,
  dampen_cycle = FALSE
)

Arguments

model

Structural time series model estimated using stsm_estimate.

y

Univariate time series of data values. May also be a 2 column data frame containing a date column.

freq

Frequency of the data (1 (yearly), 4 (quarterly), 12 (monthly), 365.25/7 (weekly), 365.25 (daily)), default is NULL and will be automatically detected

exo_obs

Matrix of exogenous variables to be used in the observation equation.

exo_state

Matrix of exogenous variables to be used in the state matrix.

ci

Confidence interval, value between 0 and 1 exclusive.

plot

Logical, whether to plot everything

plot.decomp

Logical, whether to plot the filtered historical data

n.hist

Number of historical periods to include in the forecast plot. If plot = TRUE and n.hist = NULL, defaults to 3 years.

smooth

Whether or not to use the Kalman smoother

dampen_cycle

Whether to remove oscillating cycle dynamics and smooth the cycle forecast into the trend using a sigmoid function that maintains the rate of convergence

Value

data table (or list of data tables) containing the filtered and/or smoothed series.

Examples

## Not run: 
#GDP Not seasonally adjusted
library(autostsm)
data("NA000334Q", package = "autostsm") #From FRED
NA000334Q = data.table(NA000334Q, keep.rownames = TRUE)
colnames(NA000334Q) = c("date", "y")
NA000334Q[, "date" := as.Date(date)]
NA000334Q[, "y" := as.numeric(y)]
NA000334Q = NA000334Q[date >= "1990-01-01", ]
stsm = stsm_estimate(NA000334Q)
fc = stsm_filter(stsm, y = NA000334Q, plot = TRUE)

## End(Not run)