Learn R Programming
⚠️There's a newer version (3.1.4) of this package.Take me there.

autostsm (version 1.1)

Automatic Structural Time Series Models

Description

Automatic model selection for structural time series decomposition into trend, cycle, and seasonal components using the Kalman filter. See Koopman, Siem Jan and Marius Ooms (2012) "Forecasting Economic Time Series Using Unobserved Components Time Series Models .

Copy Link

Version

Install

install.packages('autostsm')

Monthly Downloads

891

Version

1.1

License

GPL (>= 2)

Maintainer

Alex Hubbard

Last Published

January 19th, 2021

Functions in autostsm (1.1)

stsm_build_dates

Build the date sequence as a Date type
SP500

S&P 500
stsm_detect_anomalies

Detect Anomalies
stsm_detect_breaks

Detect Structural Breaks
autostsm

AutoSTSM
stsm_constraints

Set the inequality constraints for estimation
NA000334Q

US GDP Not Seasonally Adjusted
UNRATENSA

Unemployment Rate Not Seasonally Adjusted
GDP

US GDP Seasonally Adjusted
UNRATE

Unemployment Rate Seasonally Adjusted
stsm_detect_multiplicative

Detect if log transformation is best
stsm_harmonics

Get harmonics for the seasons
stsm_forecast

Kalman Filter and Forecast
stsm_prior

Return a naive model prior decomposition
stsm_detect_cycle

Detect cycle from the data
stsm_detect_frequency

Detect frequency and dates from the data
stsm_ssm

State space model
stsm_init_pars

Get initial parameter estimates for estimation
stsm_init_vals

Get initial values for the Kalman filter
stsm_detect_seasonality

Detect seasonality from the data
stsm_estimate

Trend cycle seasonal decomposition using the Kalman filter.
stsm_detect_trend

Detect trend type