AutoSTR {stR}

Automatic STR decomposition for time series data

Description

Automatically selects parameters for an STR decomposition of time series data. The time series should be of class ts or msts.

Usage

AutoSTR(
  data,
  robust = FALSE,
  gapCV = NULL,
  lambdas = NULL,
  reltol = 0.001,
  confidence = NULL,
  nsKnots = NULL,
  trace = FALSE
)

Arguments

data

A time series of class ts or msts.

robust

When TRUE, Robust STR decomposition is used. Default is FALSE.

gapCV

An optional parameter defining the length of the sequence of skipped values in the cross validation procedure.

lambdas

An optional parameter. A structure which replaces lambda parameters provided with predictors. It is used as either a starting point for the optimisation of parameters or as the exact model parameters.

reltol

An optional parameter which is passed directly to optim() when optimising the parameters of the model.

confidence

A vector of percentiles giving the coverage of confidence intervals. It must be greater than 0 and less than 1. If NULL, no confidence intervals are produced.

nsKnots

An optional vector parameter, defining the number of seasonal knots (per period) for each sesonal component.

trace

When TRUE, tracing is turned on.

Value

A structure containing input and output data. It is an S3 class STR, which is a list with the following components:

• output – contains decomposed data. It is a list of three components:

  • predictors – a list of components where each component corresponds to the input predictor. Every such component is a list containing the following:

    • data – fit/forecast for the corresponding predictor (trend, seasonal component, flexible or seasonal predictor).

    • beta – beta coefficients of the fit of the coresponding predictor.

    • lower – optional (if requested) matrix of lower bounds of confidence intervals.

    • upper – optional (if requested) matrix of upper bounds of confidence intervals.

  • random – a list with one component data, which contains residuals of the model fit.

  • forecast – a list with two components:

    • data – fit/forecast for the model.

    • beta – beta coefficients of the fit.

    • lower – optional (if requested) matrix of lower bounds of confidence intervals.

    • upper – optional (if requested) matrix of upper bounds of confidence intervals.

• input – input parameters and lambdas used for final calculations.

  • data – input data.

  • predictors - input predictors.

  • lambdas – smoothing parameters used for final calculations (same as input lambdas for STR method).

• cvMSE – optional cross validated (leave one out) Mean Squared Error.

• optim.CV.MSE – best cross validated Mean Squared Error (n-fold) achieved during minimisation procedure.

• nFold – the input nFold parameter.

• gapCV – the input gapCV parameter.

• method – always contains string "AutoSTR" for this function.

Author(s)

Alexander Dokumentov

References

Dokumentov, A., and Hyndman, R.J. (2022) STR: Seasonal-Trend decomposition using Regression, INFORMS Journal on Data Science, 1(1), 50-62. https://robjhyndman.com/publications/str/

See Also

STR

Examples

# Decomposition of a multiple seasonal time series
decomp <- AutoSTR(calls)
plot(decomp)

# Decomposition of a monthly time series
decomp <- AutoSTR(log(grocery))
plot(decomp)