To use rjwsacruncher you need to download the JWACruncher. It is available on https://github.com/jdemetra/jdemetra-app/releases or can be downloaded with the function download_cruncher():
library(rjwsacruncher)
# Directory where to save the JWSACruncher:
directory <- tempdir()
download_cruncher(directory)To configure the JWSACruncher with a portable version of Java you can use the function configure_jwsacruncher(). See ?configure_jwsacruncher for more informations.
To run the JWSACruncher you need three files:
In rjwsacruncher, there are four functions associated to run the JWSACruncher:
create_param_file() to create the “.param” parameter file;cruncher() to run the JWSACruncher on a workspace from a parameter file;cruncher_and_param() to run the ‘JWSACruncher’ on a workspace while creating the parameter file;update_workspace() to update a workspace without exporting the results.create_param_file()The parameters of the function create_param_file() are those described in the wiki of the JWSACruncher: https://github.com/jdemetra/jwsacruncher/wiki. The three most important parameters of create_param_file() are:
policy the refresh policy (see table below).| Option on JDemetra+ | Option for the JWSACruncher | Description |
|---|---|---|
| Partial concurrent adjustment -> Fixed model | current | The ARIMA model, outliers and other regression parameters are not re-identified and the values of all parameters are fixed. The transformation type remains unchanged. |
| Partial concurrent adjustment -> Estimate regression coefficients | fixedparameters (or fixed) | The ARIMA model, outliers and other regression parameters are not re-identified. The coefficients of the ARIMA model are fixed, other coefficients are re-estimated. The transformation type remains unchanged. |
| Partial concurrent adjustment -> Estimate regression coefficients + Arima parameters | parameters (by default) | The ARIMA model, outliers and other regression parameters are not re-identified. All parameters of the RegARIMA model are re-estimated. The transformation type remains unchanged. |
| Partial concurrent adjustment -> Estimate regression coefficients + Last outliers | lastoutliers | The ARIMA model, outliers (except from the outliers in the last year of the sample) and other regression parameters are not re-identified. All parameters of the RegARIMA model are re-estimated. The outliers in the last year of the sample are re-identified. The transformation type remains unchanged. |
| Partial concurrent adjustment -> Estimate regression coefficients + all outliers | outliers | The ARIMA model and regression parameters, except from outliers) are not re-identified. All parameters of the RegARIMA model are re-estimated. All outliers are re-identified. The transformation type remains unchanged. |
| Partial concurrent adjustment -> Estimate regression coefficients + Arima model | stochastic | Re-identification of the ARIMA model, outliers and regression variables, except from the calendar variables. The transformation type remains unchanged. |
| Concurrent | complete (or concurrent) | Re-identification of the whole RegARIMA model. |
matrix_item character containing the items of the matrix output. By default, the items defined in the option default_matrix_item are exported. To change it we can change the option default_matrix_item or the parameter matrix_item:# To get the default parameters
getOption("default_matrix_item")
# To change the default parameters to, for example, only export
# the information criteria:
options(default_matrix_item = c("likelihood.aic",
"likelihood.aicc",
"likelihood.bic",
"likelihood.bicc"))tsmatrix_series character containing the names of the times series to export. By default, the items defined in the option default_tsmatrix_series are exported. To change it we can change the option default_tsmatrix_series or the parameter tsmatrix_series:# To get the default parameters
getOption("default_tsmatrix_series")
# To change the default parameters to, for example, only export
# the seasonally adjusted series and its forecasts:
options(default_tsmatrix_series = c("sa", "sa_f"))For more informations, see the help of the function: ?create_param_file. Below some examples to create the parameter file:
export_dir <- tempdir()
# To create the file parameters.params in the directory export_dir with
# the refresh policy "lastoutliers" and the others default parameters:
create_param_file(dir_file_param = "D:/",
policy = "lastoutliers")
# If the options "default_matrix_item" and "default_tsmatrix_series" were
# changed to only export the information criteria, the seasonally adjusted series and its forecasts, the previous code is equivalent to:
create_param_file(dir_file_param = export_dir,
policy = "lastoutliers",
matrix_item = c("likelihood.aic", "likelihood.aicc",
"likelihood.bic", "likelihood.bicc"),
tsmatrix_series = c("sa", "sa_f"))To run the JWSACruncher with cruncher() or cruncher_and_param(), you have to specify the path to the JWSACruncher (cruncher_bin_directory parameter) and the path to the workspace (workspace parameter).
By default, the path to the JWSACruncher is the value of the option cruncher_bin_directory: therefore you only have to change this option once so that it applies to all the future running. The path to specify is the folder containing the jwsacruncher.bat file which is under the “Bin” folder of the JWSACruncher. Thus, if it is installed in D:\jdemetra-cli-2.2.2, the file jwsacruncher.bat will be under D:\jdemetra-cli-2.2.2\bin and you have to change the cruncher_bin_directory option as follows:
options(cruncher_bin_directory = "D:/jdemetra-cli-2.2.2/bin/")If no workspace is specified, a dialog box opens to select it.
The cruncher_and_param() function allows to create a temporary parameter file with create_param_file() and then run the JWSACruncher with cruncher(). In addition to the parameters available in these two functions, cruncher_and_param() allows to rename the output folder containing the workspace results so that they are equal to the names of the multi-documents displayed in the JDemetra+ software with the parameter rename_multi_documents (equals to FALSE by default). Below are some examples:
# The following code updates the workspace "workspace", that is under the folder D:/,
# with the refresh policy "lastoutliers". Others parameters are the default ones of create_param_file().
# In particular, the exported parameters are those of the options
# "default_matrix_item" options and "default_tsmatrix_series" and the results
# will be under D:/workspace/Output/.
cruncher_and_param(workspace = "D:/workspace.xml",
policy = "lastoutliers")
# Use the parameter "outpout" to change the folder that will contains the results
cruncher_and_param(workspace = "D:/workspace.xml",
output = "D:/Results/",
policy = "lastoutliers")
# To change the names of the folders containing the outputs so that they are equal
# to the names of the multi-documents displayed in JDemetra+, use the parameter
# "rename_multi_documents = TRUE". The parameter "delete_existing_file = TRUE"
# allows to delete any existing folders with the same name as the multi-documents.
cruncher_and_param(workspace = "D:/workspace.xml",
rename_multi_documents = TRUE,
delete_existing_file = TRUE,
policy = "lastoutliers")
# To see the other parameters:
?cruncher_and_param