The Problem of the SARIMA Model Selection for the Forecasting Purpose
The goal of the work is to assess the ability to identify the proper models for the time series generated by SARIMA processes with different parameter values and to analyze the accuracy of the forecasts based on the selected models. The work is based on the simulation study. To this end, a new autom...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Czech Statistical Office
2017-12-01
|
| Series: | Statistika: Statistics and Economy Journal |
| Subjects: | |
| Online Access: | https://www.czso.cz/documents/10180/45606529/32019717q4025.pdf/789389d1-539e-41da-8d41-56bd5e8573be?version=1.0 |
| _version_ | 1819267666037178368 |
|---|---|
| author | Josef Arlt Peter Trcka Markéta Arltová |
| author_facet | Josef Arlt Peter Trcka Markéta Arltová |
| author_sort | Josef Arlt |
| collection | DOAJ |
| description | The goal of the work is to assess the ability to identify the proper models for the time series generated by SARIMA processes with different parameter values and to analyze the accuracy of the forecasts based on the selected models. The work is based on the simulation study. To this end, a new automatic SARIMA modelling method is proposed. Other competing automatic SARIMA modelling procedures are applied as well and the results are compared. The important question to which the reference should be made is the relation of the magnitude of the SARIMA process parameters i. e. the size of the systematic part of the process and the ability to identify a proper model. Another issue addressed herein is the relationship between the quality of the identified model and the accuracy of forecasts achieved by its application. The simulation study leads to the results that can be generalized to most empirical analyses in various research areas. |
| first_indexed | 2024-12-23T21:20:47Z |
| format | Article |
| id | doaj.art-738f3fbe14b045dca1ce7d6ce61a17a1 |
| institution | Directory Open Access Journal |
| issn | 0322-788X 1804-8765 |
| language | English |
| last_indexed | 2024-12-23T21:20:47Z |
| publishDate | 2017-12-01 |
| publisher | Czech Statistical Office |
| record_format | Article |
| series | Statistika: Statistics and Economy Journal |
| spelling | doaj.art-738f3fbe14b045dca1ce7d6ce61a17a12022-12-21T17:30:46ZengCzech Statistical OfficeStatistika: Statistics and Economy Journal0322-788X1804-87652017-12-019742532The Problem of the SARIMA Model Selection for the Forecasting PurposeJosef Arlt0Peter Trcka1Markéta Arltová2University of Economics, Prague, Czech RepublicUniversity of Economics, Prague, Czech RepublicUniversity of Economics, Prague, Czech RepublicThe goal of the work is to assess the ability to identify the proper models for the time series generated by SARIMA processes with different parameter values and to analyze the accuracy of the forecasts based on the selected models. The work is based on the simulation study. To this end, a new automatic SARIMA modelling method is proposed. Other competing automatic SARIMA modelling procedures are applied as well and the results are compared. The important question to which the reference should be made is the relation of the magnitude of the SARIMA process parameters i. e. the size of the systematic part of the process and the ability to identify a proper model. Another issue addressed herein is the relationship between the quality of the identified model and the accuracy of forecasts achieved by its application. The simulation study leads to the results that can be generalized to most empirical analyses in various research areas.https://www.czso.cz/documents/10180/45606529/32019717q4025.pdf/789389d1-539e-41da-8d41-56bd5e8573be?version=1.0SARIMAsimulationidentification of modelforecasting |
| spellingShingle | Josef Arlt Peter Trcka Markéta Arltová The Problem of the SARIMA Model Selection for the Forecasting Purpose Statistika: Statistics and Economy Journal SARIMA simulation identification of model forecasting |
| title | The Problem of the SARIMA Model Selection for the Forecasting Purpose |
| title_full | The Problem of the SARIMA Model Selection for the Forecasting Purpose |
| title_fullStr | The Problem of the SARIMA Model Selection for the Forecasting Purpose |
| title_full_unstemmed | The Problem of the SARIMA Model Selection for the Forecasting Purpose |
| title_short | The Problem of the SARIMA Model Selection for the Forecasting Purpose |
| title_sort | problem of the sarima model selection for the forecasting purpose |
| topic | SARIMA simulation identification of model forecasting |
| url | https://www.czso.cz/documents/10180/45606529/32019717q4025.pdf/789389d1-539e-41da-8d41-56bd5e8573be?version=1.0 |
| work_keys_str_mv | AT josefarlt theproblemofthesarimamodelselectionfortheforecastingpurpose AT petertrcka theproblemofthesarimamodelselectionfortheforecastingpurpose AT marketaarltova theproblemofthesarimamodelselectionfortheforecastingpurpose AT josefarlt problemofthesarimamodelselectionfortheforecastingpurpose AT petertrcka problemofthesarimamodelselectionfortheforecastingpurpose AT marketaarltova problemofthesarimamodelselectionfortheforecastingpurpose |