簡易檢索 / 詳目顯示

回結果列表
研究生： 李童宇
Lee, Tung-Yu
論文名稱： 擁有少量歷史資料的新產品銷售預測：比較時間序列配對方法與利用時間序列群集預測的策略差異
Sales Forecasting for New Items with Short Histories: Comparing Strategies for Time­-Series Matching and Forecasting using Time­-Series Clusters
指導教授： 雷松亞
Ray, Soumya
口試委員: 林福仁
Lin, Fu-Ren
徐茉莉
Shmueli, Galit
學位類別： 碩士
Master
系所名稱： 科技管理學院 - 服務科學研究所
Institute of Service Science
論文出版年： 2017
畢業學年度： 105
語文別： 英文
論文頁數： 64
中文關鍵詞： 銷售預測 、少量的歷史資料 、時間序列群集分析 、原型 、時間序列配對分法
外文關鍵詞： Sales Forecasting, Short­-history data, Time­-series Clustering, Prototypes, Time­-series Matching
相關次數： 點閱：1  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 時間序列預測是一種相當受歡迎的方法，並且被廣泛運用在許多不同領域。然而 ，在使用常見的時間序列預測演算法時，會經常遇到一個嚴重的問題:歷史資料不 足。此研究提出了一個三步驟的方法，可以自動化的為歷史資料不足的產品作出預 測。此三步驟的方法，是由時間序列群集分法、時間序列配對方法和預測方法所組成 ，在各個步驟中，我們測試了不同的策略，因此總共12種策略組合在此研究中被測 試。我們使用的是真實的銷售數據，此數據是由台灣提供平台服務的公司iCHEF所提 供。此研究我們比較了不同策略組合對於預測能力的差異，並發現不同的時間序列配 對方法對於預測能力有顯著的影響，除此之外，當要預測歷史資料不足的產品時，使 用相似的產品的歷史資料，也是很有用的資訊。

    Time­-series forecasting is a popular technique and is used for different purposes in many fields. However, one big problem exists in many common forecasting algorithms is the short­-history problem. Our study presents a 3­-step method that can automatically provide forecasts for items with very short histories. This 3­-step method composes of time­-series clustering, time­-series matching, and forecasting methods. There are several different strategies in each step, so 12 combinations are conducted in our study. We compare the forecasting performance of different combinations using real sales data from iCHEF, a Taiwanese platform company. The results show that time­-series matching methods influence forecasting performance significantly. Besides, similar historical data is useful information while forecasting for items with short histories.

    摘要 1 Abstract 2 Acknowledgement 3 Table of Contents 4 List of Figures 6 List of Tables 7 Chapter 1: Introduction 8 Chapter 2: Study Context 11 2.1. Forecasting analytics with iCHEF platform 11 2.2. Short histories problem at iCHEF 12 Chapter 3: A Short History of the Short History Problem 13 3.1. Related works on solving short histories and no history problem 13 3.2. Neural clustering and classification (NCC) for forecasting with no history 14 Chapter 4: Methodology 17 4.1. Data collection and pre­processing 17 4.2. Proposed Method 19 Chapter 5: Benchmarking Design 42 5.1. Study Combinations 42 5.2. Study Implementation in R 43 Chapter 6: Results 45 6.1 Results of Step 1a: SOM clustering 45 6.2 Results of Step 1b: Hierarchical clustering on SOM nodes 46 6.3 Comparing results of Step 2: Matching Time­series 48 6.4 Forecasting performance (Step 3): proposed strategies against naive 52 Chapter 7: Discussion of Findings 56 Chapter 8: Limitations 59 Chapter 9: Conclusion 61 References 63

    Bass, F. M. (1969). A new product growth for model consumer durables. Management science, 15(5), 215­-227.

    Bezdek, J. C., & Pal, N. R. (1998). Some new indexes of cluster validity. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 28(3), 301­-315.

    Dharun, V. S., & Karnan, M. (2012). Voice and speech recognition for tamil words and numerals. International Journal of Modern Engineering Research (IJMER), 2(5), 3406­-3414.

    Garcı́a, H. L., & González, I. M. (2004). Self­organizing map and clustering for wastewater treatment monitoring. Engineering Applications of Artificial Intelligence, 17(3), 215­-225.

    Giorgino, T. (2009). Computing and visualizing dynamic time warping alignments in R: the dtw package. Journal of statistical Software, 31(7), 1-­24.

    Han, J., Pei, J., & Kamber, M. (2006). Data Mining. Morgan Kaufmann.

    Jung, Y., Park, H., Du, D. Z., & Drake, B. L. (2003). A decision criterion for the optimal number of clusters in hierarchical clustering. Journal of Global Optimization, 25(1), 91­-111.

    Kohonen, T. (1998). The self­organizing map. Neurocomputing, 21(1), 1­-6.

    Kiang, M. Y. (2001). Extending the Kohonen self­organizing map networks for clustering
    analysis. Computational Statistics & Data Analysis, 38(2), 161­-180.

    Morrison, J. (1996). How to use diffusion models in new product forecasting. The Journal of
    Business Forecasting, 15(2), 6.

    Shmueli, G., Patel, N. R., & Bruce, P. C. (2011). Data mining for business intelligence:
    Concepts, techniques, and applications in Microsoft Office Excel with XLMiner. John Wiley and Sons.

    Shmueli, G., & Lichtendahl, K. C. (2016). Practical Time Series Forecasting with R: A Hands­On Guide. Axelrod Schnall Publishers.

    Thomassey, S., & Happiette, M. (2007). A neural clustering and classification system for sales forecasting of new apparel items. Applied Soft Computing, 7(4), 1177­-1187.

    Tibshirani, R., Walther, G., & Hastie, T. (2001). Estimating the number of clusters in a data set via the gap statistic. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 63(2), 411­-423.

    QR CODE