WEKO3
アイテム
Car Sales Forecasting Methods for One Model: Current Status and Challenges
https://kindai.repo.nii.ac.jp/records/2000640
https://kindai.repo.nii.ac.jp/records/2000640802338b2-eb31-467a-98ad-56288fed53b9
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
AA12517632-20231200-0077.pdf
|
|
| Item type | 紀要論文 / departmental bulletin paper(1) | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2024-01-19 | |||||||||||||||||
| タイトル | ||||||||||||||||||
| タイトル | Car Sales Forecasting Methods for One Model: Current Status and Challenges | |||||||||||||||||
| 言語 | en | |||||||||||||||||
| その他のタイトル | ||||||||||||||||||
| その他のタイトル | 1車種における自動車販売予測アプローチ：現状と課題 | |||||||||||||||||
| 言語 | ja | |||||||||||||||||
| 作成者 |
Yomono, Tomonari
× Yomono, Tomonari
× Kataoka, Takayuki
|
|||||||||||||||||
| 言語 | ||||||||||||||||||
| 言語 | eng | |||||||||||||||||
| キーワード | ||||||||||||||||||
| 主題 | Time Series, Demand Forecast, Deep Learning, Statistical Methods | |||||||||||||||||
| 内容記述 | ||||||||||||||||||
| 内容記述タイプ | Abstract | |||||||||||||||||
| 内容記述 | The traditional car sales forecasting methods have some problems to make production plans without depending on experience or intuition. Almost all previous papers have been focused on forecasting total sales volume for automobile markets or manufacturers. However, micro forecasting methods for one model are required in the actual car manufacturer production plans. In addition, the accuracy and stability of forecasts differ greatly depending on the car model (Sedan, SUV and Wagon, etc.). In this paper, the relationship between automobile sales data and forecasting methods is forecasted and investigated for the sales volume of one type car model. In experimental results, deep learning methods and statistical forecasting methods for sales data of two type car models are considered. As the results, statistical methods can lead high forecasting accuracy in many periods, although it shows a significant decrease in the stability of the forecasting accuracy in some periods. On the other hand, one of deep learning methods, GRU, can lead relatively stable forecasting results for small-scale data. | |||||||||||||||||
| 言語 | en | |||||||||||||||||
| 内容記述 | ||||||||||||||||||
| 内容記述タイプ | TableOfContents | |||||||||||||||||
| 内容記述 | III. 論文集 3. 再録論文 |
|||||||||||||||||
| 言語 | ja | |||||||||||||||||
| 出版者 | ||||||||||||||||||
| 出版者 | 近畿大学次世代基盤技術研究所 | |||||||||||||||||
| 言語 | ja | |||||||||||||||||
| 資源タイプ | ||||||||||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||||||||||||||
| 資源タイプ | departmental bulletin paper | |||||||||||||||||
| 出版タイプ | ||||||||||||||||||
| 出版タイプ | AM | |||||||||||||||||
| 出版タイプResource | http://purl.org/coar/version/c_ab4af688f83e57aa | |||||||||||||||||
| 収録物識別子 | ||||||||||||||||||
| 収録物識別子タイプ | PISSN | |||||||||||||||||
| 収録物識別子 | 21858802 | |||||||||||||||||
| 書誌情報 |
ja : 近畿大学次世代基盤技術研究所報告 en : Annual Report of Fundamental Technology for Next Generation Research Institute, Kindai University 号 14, p. 77-82, 発行日 2023-12 |
|||||||||||||||||
