컨텐츠 시작
학술대회/행사
초록검색
| 제출번호(No.) | 0482 |
|---|---|
| 분류(Section) | Poster Session |
| 분과(Session) | Probability / Stochastic Process / Statistics (SS-12) |
| 영문제목 (Title(Eng.)) |
[CANCELLED] Two levels model based on ARIMAX and Regression method for forecasting retail sales data with holiday effects |
| 저자(Author(s)) |
Suhartono1, Muhammad Hisyam Lee2 Department of Statistics, Institut Teknologi Sepuluh Nopember, Indonesia1, Department of Mathematics, Universiti Teknologi Malaysia2 |
| 초록본문(Abstract) | The objective of this research is to propose a calendar variation model for forecasting retail sales data with holiday effect, namely the Eid ul-Fitr effect. The proposed model was developed based on two methods, namely ARIMAX and regression methods. This two levels model was built by using ARIMAX for the first level and regression for the second level. Monthly men’s jeans and women's trousers sales in a retail company were used as case study. In general, two levels of calendar variation model yielded two models, namely the first model to reconstruct the sales pattern that already occurred, and the second model to forecast the effect of increasing sales due to Eid ul-Fitr that affected sales at the same and the previous months. The results of forecast accuracy comparison showed that the proposed two level calendar variation model yielded better forecast compared to the seasonal ARIMA model and Neural Networks. |
| 분류기호 (MSC number(s)) |
91B84 |
| 키워드(Keyword(s)) | two levels, ARIMAX, regression, calendar effect, holiday |
| 강연 형태 (Language of Session (Talk)) |
English |