學術產出-Theses
Article View/Open
Publication Export
-
題名 提高銷售預測準確性:判斷式、Prophet及混合預測方法之比較分析
Enhancing sales forecasting accuracy: a comparative analysis of judgmental, prophet, and hybrid forecasting approaches作者 卡西歐
Finotti, Cassio C.貢獻者 莊皓鈞
Chuang, Howard
卡西歐
Cassio C. Finotti關鍵詞 預測
預測模型
判斷性預測
銷售數據
時間序列分析
Forecasting
Prophet model
Judgmental forecasts
Sales data
Time-series analysis日期 2023 上傳時間 6-Jul-2023 16:34:12 (UTC+8) 摘要 .
This thesis explores how Prophet, a time-series model by Meta, can enhance judgmental forecasts for predicting the monthly demand of ten products from a single customer of a B2B manufacturing company. The dataset spans from 2018-2022, providing monthly sales data with 2022 as the focus. Forecast accuracy is assessed using the Cumulative Forecast Error (CFE) method. Results show that the Prophet model excels judgmental forecasts in 6 out of 10 products, and a Hybrid approach of incorporating judgmental forecasts as regressors improve performance, outperforming in 8 out of 10 products. The findings show the benefits of integrating advanced statistical models like Prophet into business forecasting processes to mitigate over and underforecasting and boost accuracy. The study outlines limitations and future research opportunities, such as expanding datasets, exploring new comparison metrics, and periodically updating the Prophet model. Practical implications discuss challenges and benefits of statistical forecasting models, Prophet’s accessibility, and the need to counter underforecasting and overforecasting. By harnessing new technologies, businesses can enhance operations and improve demand forecasting accuracy. This thesis highlights the potential of merging statistical models like Prophet with judgmental forecasts and proposes areas for further exploration to refine these models’ effectiveness in business contexts.參考文獻 Reference1. Chen, H., Frank, M. Z., & Wu, O. Q. (2005). What Actually Happened to the Inventories of American Companies Between 1981 and 2000? Management Science, 51(7), 1015-1031.2. De Livera, A., Hyndman, R. J., & Snyder, R. (2010). Forecasting Time Series With Complex Seasonal Patterns Using Exponential Smoothing. Journal of the American Statistical Association, 106, 1513-1527.3. Facebook. (n.d.). Quick start — Prophet 1.0 documentation. Retrieved March 31, 2023, from https://facebook.github.io/prophet/docs/quick_start.html4. Fildes, R., & Goodwin, P. (2007). Against your better judgment? How organizations can improve their use of management judgment in forecasting. Interfaces, 37(6), 570–576. http://www.jstor.org/stable/201415475. Goodwin, P. (2002). Integrating management judgment and statistical methods to improve short-term forecasts. Omega, 30(2), 127-135.6. Helmer, O. (1983). Looking forward: a guide to future research. Sage Publications, Inc.7. Hsu, C. C., & Sandford, B. A. (2007). The Delphi technique: making sense of consensus. Practical Assessment, Research, and Evaluation, 12(1), 10.8. Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: principles and practice. OTexts.9. Kutzkov, K. (2023, April 19). ARIMA vs Prophet vs LSTM for Time Series Prediction [web log]. Retrieved April 22, 2023, from https://neptune.ai/blog/arima-vs-prophet-vs-lstm.10. Lawrence, M., Goodwin, P., O`Connor, M., & Önkal, D. (2006). Judgmental forecasting: A review of progress over the last 25 years. International Journal of Forecasting, 22(3).11. Menculini, L., Marini, A., Proietti, M., Garinei, A., Bozza, A., Moretti, C., & Marconi, M. (2021). Comparing Prophet and Deep Learning to ARIMA in Forecasting Wholesale Food Prices. Forecasting, 3, 644–662.12. Moon, M. A., Mentzer, J. T., & Smith, C. D. (2003). Researching Sales Forecasting Practice: Commentaries and authors` response on "Conducting a Sales Forecasting Audit." International Journal of Forecasting, 19, 27-42.13. Rowe, G., & Wright, G. (1999). The Delphi technique as a forecasting tool: issues and analysis. International Journal of Forecasting, 15(4), 353-375.14. Smolic, H. (2022, September 1). Sales Forecasting: How to Apply Machine Learning. Graphite Note. Retrieved May 10, 2023, from https://graphite-note.com/machine-learning-sales-forecasting15. Taylor, S. J., & Letham, B. (2017). Forecasting at Scale. The American Statistician, 72(1).16. Van der Heijden, K. (2005). Scenarios: The Art of Strategic Conversation. John Wiley & Sons.17. Van Houdt, G., Mosquera, C., & Nápoles, G. (2020). A review on the long short-term memory model. Artificial Intelligence Review, 53, 5929–5955.18. Wright, G., & Goodwin, P. (2009). Decision making and planning under low levels of predictability: Enhancing the scenario method. International Journal of Forecasting, 25(4), 813-825. 描述 碩士
國立政治大學
國際經營管理英語碩士學位學程(IMBA)
110933046資料來源 http://thesis.lib.nccu.edu.tw/record/#G0110933046 資料類型 thesis dc.contributor.advisor 莊皓鈞 zh_TW dc.contributor.advisor Chuang, Howard en_US dc.contributor.author (Authors) 卡西歐 zh_TW dc.contributor.author (Authors) Cassio C. Finotti en_US dc.creator (作者) 卡西歐 zh_TW dc.creator (作者) Finotti, Cassio C. en_US dc.date (日期) 2023 en_US dc.date.accessioned 6-Jul-2023 16:34:12 (UTC+8) - dc.date.available 6-Jul-2023 16:34:12 (UTC+8) - dc.date.issued (上傳時間) 6-Jul-2023 16:34:12 (UTC+8) - dc.identifier (Other Identifiers) G0110933046 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/145804 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 國際經營管理英語碩士學位學程(IMBA) zh_TW dc.description (描述) 110933046 zh_TW dc.description.abstract (摘要) . zh_TW dc.description.abstract (摘要) This thesis explores how Prophet, a time-series model by Meta, can enhance judgmental forecasts for predicting the monthly demand of ten products from a single customer of a B2B manufacturing company. The dataset spans from 2018-2022, providing monthly sales data with 2022 as the focus. Forecast accuracy is assessed using the Cumulative Forecast Error (CFE) method. Results show that the Prophet model excels judgmental forecasts in 6 out of 10 products, and a Hybrid approach of incorporating judgmental forecasts as regressors improve performance, outperforming in 8 out of 10 products. The findings show the benefits of integrating advanced statistical models like Prophet into business forecasting processes to mitigate over and underforecasting and boost accuracy. The study outlines limitations and future research opportunities, such as expanding datasets, exploring new comparison metrics, and periodically updating the Prophet model. Practical implications discuss challenges and benefits of statistical forecasting models, Prophet’s accessibility, and the need to counter underforecasting and overforecasting. By harnessing new technologies, businesses can enhance operations and improve demand forecasting accuracy. This thesis highlights the potential of merging statistical models like Prophet with judgmental forecasts and proposes areas for further exploration to refine these models’ effectiveness in business contexts. en_US dc.description.tableofcontents TABLE OF CONTENTS1. Introduction: Traditional Approaches to Sales Forecasting 11.1. Limitations of Statistical and Judgmental Forecasts 31.2. Company background and Forecasting Process 41.3. The importance of Managing Overforecasting and Underforecasting in Sales Forecasting 71.4. The Potential of Prophet Model in Improving Forecasting Accuracy and Efficiency 82. Literature Review 102.1. Judgmental Forecasting Techniques 102.2. Statistical Forecasting Techniques 122.3. Comparison between Prophet, ARIMA and Neural Network Models 142.4. Combining Judgmental Forecasting with Statistical Forecasting 152.5. Overforecasting and Underforecasting 162.6. Conclusion of the Literature Review 163. Methodology 183.1. Data Collection and Preprocessing 183.2. Forecasting Models 193.3. Model Comparison 214. Results and Discussion 244.1. Comparison of Performance for Judgmental, Prophet, and Hybrid Forecasts 244.2. Analysis and Comparison of Forecasting Methods 354.3. Implications 425. Conclusion 445.1. Summary of Findings 445.2. Limitations and Future Research 465.3. Practical Implications for Businesses 47Reference 50 List of Figures and TablesFigure 1: Sales Forecasting and Production Process Flowchart 6Figure 2: Comparison of results for Part # 1 25Figure 3: Comparison of results for Part # 2 26Figure 4: Comparison of results for Part # 3 27Figure 5: Comparison of Results for Part #4 28Figure 6: Comparison of results for Part #5 29Figure 7: Comparison of results for Part #6 30Figure 8: Comparison of results for Part #7 31Figure 9: Comparison of results for Part #8 32Figure 10: Comparison of results for Part #9 33Figure 11: Comparison of results for Part #10 34Figure 12: Comparison of Cumulative Forecast Error for Products using Judgmental, Prophet, and Hybrid Models 41Table 1: Comparison of Cumulative Forecast Errors (CFE) for Judgmental Forecast, Prophet Forecast, and Prophet Forecast with Judgmental Forecast as Regressor………………………………………………………………………………………….…33 zh_TW dc.format.extent 1268907 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0110933046 en_US dc.subject (關鍵詞) 預測 zh_TW dc.subject (關鍵詞) 預測模型 zh_TW dc.subject (關鍵詞) 判斷性預測 zh_TW dc.subject (關鍵詞) 銷售數據 zh_TW dc.subject (關鍵詞) 時間序列分析 zh_TW dc.subject (關鍵詞) Forecasting en_US dc.subject (關鍵詞) Prophet model en_US dc.subject (關鍵詞) Judgmental forecasts en_US dc.subject (關鍵詞) Sales data en_US dc.subject (關鍵詞) Time-series analysis en_US dc.title (題名) 提高銷售預測準確性:判斷式、Prophet及混合預測方法之比較分析 zh_TW dc.title (題名) Enhancing sales forecasting accuracy: a comparative analysis of judgmental, prophet, and hybrid forecasting approaches en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) Reference1. Chen, H., Frank, M. Z., & Wu, O. Q. (2005). What Actually Happened to the Inventories of American Companies Between 1981 and 2000? Management Science, 51(7), 1015-1031.2. De Livera, A., Hyndman, R. J., & Snyder, R. (2010). Forecasting Time Series With Complex Seasonal Patterns Using Exponential Smoothing. Journal of the American Statistical Association, 106, 1513-1527.3. Facebook. (n.d.). Quick start — Prophet 1.0 documentation. Retrieved March 31, 2023, from https://facebook.github.io/prophet/docs/quick_start.html4. Fildes, R., & Goodwin, P. (2007). Against your better judgment? How organizations can improve their use of management judgment in forecasting. Interfaces, 37(6), 570–576. http://www.jstor.org/stable/201415475. Goodwin, P. (2002). Integrating management judgment and statistical methods to improve short-term forecasts. Omega, 30(2), 127-135.6. Helmer, O. (1983). Looking forward: a guide to future research. Sage Publications, Inc.7. Hsu, C. C., & Sandford, B. A. (2007). The Delphi technique: making sense of consensus. Practical Assessment, Research, and Evaluation, 12(1), 10.8. Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: principles and practice. OTexts.9. Kutzkov, K. (2023, April 19). ARIMA vs Prophet vs LSTM for Time Series Prediction [web log]. Retrieved April 22, 2023, from https://neptune.ai/blog/arima-vs-prophet-vs-lstm.10. Lawrence, M., Goodwin, P., O`Connor, M., & Önkal, D. (2006). Judgmental forecasting: A review of progress over the last 25 years. International Journal of Forecasting, 22(3).11. Menculini, L., Marini, A., Proietti, M., Garinei, A., Bozza, A., Moretti, C., & Marconi, M. (2021). Comparing Prophet and Deep Learning to ARIMA in Forecasting Wholesale Food Prices. Forecasting, 3, 644–662.12. Moon, M. A., Mentzer, J. T., & Smith, C. D. (2003). Researching Sales Forecasting Practice: Commentaries and authors` response on "Conducting a Sales Forecasting Audit." International Journal of Forecasting, 19, 27-42.13. Rowe, G., & Wright, G. (1999). The Delphi technique as a forecasting tool: issues and analysis. International Journal of Forecasting, 15(4), 353-375.14. Smolic, H. (2022, September 1). Sales Forecasting: How to Apply Machine Learning. Graphite Note. Retrieved May 10, 2023, from https://graphite-note.com/machine-learning-sales-forecasting15. Taylor, S. J., & Letham, B. (2017). Forecasting at Scale. The American Statistician, 72(1).16. Van der Heijden, K. (2005). Scenarios: The Art of Strategic Conversation. John Wiley & Sons.17. Van Houdt, G., Mosquera, C., & Nápoles, G. (2020). A review on the long short-term memory model. Artificial Intelligence Review, 53, 5929–5955.18. Wright, G., & Goodwin, P. (2009). Decision making and planning under low levels of predictability: Enhancing the scenario method. International Journal of Forecasting, 25(4), 813-825. zh_TW