Demetra Components#

Below is a list of the available components and their descriptions:

Original and Forecasted Series#

  • y : Original series

  • y_f : Forecasts of the original series

  • y_ef : Standard errors of the forecasts of the original

  • y_c : Interpolated series

  • yc_f : Forecasts of the interpolated series

  • yc_ef : Standard errors of the forecasts of the interpolated

Linearized and Corrected Series#

  • y_lin : Linearised series (not transformed)

  • l : Linearised series (transformed)

  • ycal : Series corrected for calendar effects

  • ycal_f : Forecasts of the series corrected for calendar effects

  • l_f : Forecasts of the linearised series

  • l_b : Backcasts of the linearised series

Trend and Seasonally Adjusted Components#

  • t : Trend (including deterministic effects)

  • t_f : Forecasts of the trend

  • sa : Seasonally adjusted series (including deterministic effects)

  • sa_f : Forecasts of the seasonally adjusted series

Seasonal, Irregular, and Deterministic Components#

  • s : Seasonal component (including deterministic effects)

  • s_f : Forecasts of the seasonal component

  • i : Irregular component (including deterministic effects)

  • i_f : Forecasts of the irregular component

  • det : All deterministic effects

  • det_f : Forecasts of the deterministic effects

Calendar and Holiday Effects#

  • cal : Calendar effects

  • cal_f : Forecasts of the calendar effects

  • tde : Trading day effect

  • tde_f : Forecasts of the trading day effect

  • mhe : Moving holidays effects

  • mhe_f : Forecasts of the moving holidays effects

  • ee : Easter effect

  • ee_f : Forecasts of the Easter effect

  • omhe : Other moving holidays effects

  • omhe_f : Forecasts of the other moving holidays effects

Outliers and Regression Effects#

  • out : All outliers effects

  • out_f : Forecasts of all outliers effects

  • out_i : Outliers effects related to irregular (AO, TC)

  • out_i_f : Forecasts of outliers effects related to irregular (TC)

  • out_t : Outliers effects related to trend (LS)

  • out_t_f : Forecasts of outliers effects related to trend (LS)

  • out_s : Outliers effects related to seasonal (SO)

  • out_s_f : Forecasts of outliers effects related to seasonal (SO)

  • reg : All other regression effects

  • reg_f : Forecasts of all other regression effects

  • reg_i : Regression effects related to irregular

  • reg_i_f : Forecasts of regression effects related to irregular

  • reg_t : Regression effects related to trend

  • reg_t_f : Forecasts of regression effects related to trend

  • reg_s : Regression effects related to seasonal

  • reg_s_f : Forecasts of regression effects related to seasonal

  • reg_sa : Regression effects related to seasonally adjusted series

  • reg_sa_f : Forecasts of regression effects related to seasonally adjusted

  • reg_y : Separate regression effects

  • reg_y_f : Forecasts of separate regression effects

RegARIMA Model and Decomposition Components#

  • fullresiduals : Full residuals of the RegARIMA model

  • decomposition.y_lin : Linearised series used as input in the decomposition

  • decomposition.y_lin_f : Forecast of the linearised series used as input

  • decomposition.t_lin : Trend produced by the decomposition

  • decomposition.t_lin_f : Forecasts of the trend produced by the decomposition

  • decomposition.s_lin : Seasonal component produced by the decomposition

  • decomposition.s_lin_f : Forecasts of the seasonal component produced by the decomposition

  • decomposition.i_lin : Irregular component produced by the decomposition

  • decomposition.i_lin_f : Forecasts of the irregular component produced by the decomposition

  • decomposition.sa_lin : Seasonally adjusted series produced by the decomposition

  • decomposition.sa_lin_f : Forecasts of the seasonally adjusted series produced by the decomposition

  • decomposition.si_lin : Seasonal-Irregular component produced by the decomposition

Benchmarking and Other Components#

  • lambda decomposition_a-tables_a1 : For X-13ARIMA-SEATS only. Series

  • benchmarking_result : Benchmarked seasonally adjusted series

  • benchmarking_target : Target for the benchmarking