RegARIMA model, pre-adjustment in X13
Arguments
- ts
an univariate time series.
- spec
the model specification. Can be either the name of a predefined specification or a user-defined specification.
- context
list of external regressors (calendar or other) to be used for estimation
- userdefined
a vector containing additional output variables (see
x13_dictionary()).
Value
the regarima() function returns a list with the results ("JD3_REGARIMA_RSLTS" object), the estimation specification and the result specification, while regarima_fast() is a faster function that only returns the results.
Examples
y = rjd3toolkit::ABS$X0.2.09.10.M
sp = regarima_spec("rg5c")
sp = rjd3toolkit::add_outlier(sp,
type = c("AO"), c("2015-01-01", "2010-01-01"))
regarima_fast(y, spec = sp)
#> Log-transformation: yes
#> SARIMA model: (0,1,1) (1,1,1)
#>
#> Coefficients
#> Estimate Std. Error T-stat
#> theta(1) -0.81606 0.06959 -11.73
#> bphi(1) -0.43734 0.02661 -16.44
#> btheta(1) -0.82509 0.04481 -18.41
#>
#> Regression model:
#> Estimate Std. Error T-stat
#> monday -0.008747 0.003287 -2.661
#> tuesday 0.004488 0.003314 1.354
#> wednesday -0.001471 0.003294 -0.447
#> thursday 0.013886 0.003325 4.176
#> friday -0.001944 0.003325 -0.585
#> saturday 0.015368 0.003304 4.651
#> easter 0.051130 0.006621 7.723
#> AO (2010-01-01) 0.035349 0.028796 1.228
#> AO (2015-01-01) -0.020385 0.028885 -0.706
#> TC (2000-06-01) 0.162169 0.026510 6.117
#> AO (2000-07-01) -0.306536 0.032095 -9.551
#> Number of observations: 425
#> Number of effective observations: 412
#> Number of parameters: 15
#>
#> Loglikelihood: 795.8429
#> Adjusted loglikelihood: -2071.784
#>
#> Standard error of the regression (ML estimate): 0.03478056
#> AIC: 4173.568
#> AICC: 4174.78
#> BIC: 4233.883
#>
sp = rjd3toolkit::set_transform(
rjd3toolkit::set_tradingdays(
rjd3toolkit::set_easter(sp, enabled = FALSE),
option = "workingdays"
),
fun = "None"
)
regarima_fast(y, spec = sp)
#> Log-transformation: no
#> SARIMA model: (3,1,1) (0,1,1)
#>
#> Coefficients
#> Estimate Std. Error T-stat
#> phi(1) 0.16557 0.06461 2.563
#> phi(2) 0.08780 0.06435 1.364
#> phi(3) -0.11287 0.06047 -1.867
#> theta(1) -0.86076 0.03927 -21.918
#> btheta(1) -0.22918 0.05193 -4.413
#>
#> Regression model:
#> Estimate Std. Error T-stat
#> td 1.0015 0.8098 1.237
#> lp 29.6392 11.5469 2.567
#> AO (2010-01-01) 37.0476 34.4300 1.076
#> AO (2015-01-01) 27.9946 35.0234 0.799
#> AO (2000-06-01) 199.7536 34.5756 5.777
#> AO (2000-07-01) -194.6887 34.6240 -5.623
#> LS (2005-04-01) -82.3062 17.4210 -4.725
#> LS (2015-07-01) 81.3334 18.0924 4.495
#> Number of observations: 425
#> Number of effective observations: 412
#> Number of parameters: 14
#>
#> Loglikelihood: -2159.902
#> Standard error of the regression (ML estimate): 45.632
#> AIC: 4347.804
#> AICC: 4348.862
#> BIC: 4404.099
#>
sp = rjd3toolkit::set_outlier(sp, outliers.type = c("AO"))
regarima_fast(y, spec = sp)
#> Log-transformation: no
#> SARIMA model: (3,1,1) (0,1,1)
#>
#> Coefficients
#> Estimate Std. Error T-stat
#> phi(1) 0.11808 0.09633 1.226
#> phi(2) 0.03364 0.09199 0.366
#> phi(3) -0.15061 0.08002 -1.882
#> theta(1) -0.83611 0.07714 -10.838
#> btheta(1) -0.24114 0.05412 -4.456
#>
#> Regression model:
#> Estimate Std. Error T-stat
#> td 1.0012 0.8028 1.247
#> lp 30.8981 11.7820 2.622
#> AO (2010-01-01) 37.2796 35.1800 1.060
#> AO (2015-01-01) 6.5622 35.1307 0.187
#> AO (2000-06-01) 194.6157 35.1753 5.533
#> AO (2000-07-01) -201.1923 35.2286 -5.711
#> AO (2005-04-01) -150.2768 35.1047 -4.281
#> Number of observations: 425
#> Number of effective observations: 412
#> Number of parameters: 13
#>
#> Loglikelihood: -2169.985
#> Standard error of the regression (ML estimate): 46.77874
#> AIC: 4365.971
#> AICC: 4366.885
#> BIC: 4418.244
#>