Abstract
In this paper, an ARIMA based demand forecasting program has been implemented by utilizing public data on G2B e-commerce at public procurement service. Recently, there has been growing interest in research and use of public data. However, it is found that few studies proposed systematic procedures for collecting, processing, and analysing the public data. There also have been limitations on grasping the attributes of the public data, extracting them for researchers’ purpose and applying the right methodologies. This study presents a series of procedures for redefining and extracting the public data, and attempts to forecast the market sizes in each segment using the ARIMA model. This paper also sheds light on utilizing the public data in the area of demand forecasting by programming the whole processes where unit root test, model identification, model estimation, and model diagnostic checking are automatically to be done. As a result of the analysis, MAPE values fall into 3.90%~24.47%, which shows that the ARIMA model implemented in this paper is working well.
| Translated title of the contribution | Demand Forecasting for G2B E-commerce Using Public Data : A Case Study of Public Procurement Service |
|---|---|
| Original language | Korean |
| Pages (from-to) | 113-121 |
| Number of pages | 9 |
| Journal | 한국정보기술학회논문지 |
| Volume | 12 |
| Issue number | 10 |
| DOIs | |
| State | Published - Oct 2014 |