臺灣博碩士論文加值系統

本研究深入探討了美國職業棒球大聯盟的動態領域，並運用廣義估計方程式來預測短期觀眾參與情況。
研究聚焦於短期需求與比賽結果不確定性之間的複雜互動，旨在深入解析影響觀眾偏好的多重因素。
透過廣義估計方程式的應用，我們分析了隊伍表現、比賽動態以及觀眾出席之間的多面關係。
這裡的「比賽動態」不僅指涉單場比賽內部的變化和影響，更包括了比賽之間的互動效應，揭示了其對觀眾參與的重要性及影響程度。
相對於傳統的廣義線性模型，廣義估計方程式更有效地解決了模型忽略不同隊伍和比賽間相關性的問題，特別是在捕捉棒球觀眾體驗的複雜性方面至為重要。
本研究利用了 2019 年、2021 年和 2022 年各支隊伍最後 40 場主場比賽的數據，深入探索了觀眾興趣的特性。
這項研究不僅為運動分析領域的進步貢獻力量，同時為希望在職業體育不斷變化的環境中優化球迷參與策略的棒球相關利益者提供了實用的見解。

This study explores the dynamic realm of Major League Baseball (MLB) using Generalized Estimating Equations (GEE) to predict audience engagement.
Focused on the uncertainty of game outcomes, our research aims to uncover the intricate factors influencing viewer preferences.
By employing GEE, we analyze the multifaceted relationships among team performance, game dynamics, and audience attendance.
Unlike traditional generalized linear models, GEE addresses issues such as the correlation between different teams and games, proving essential in capturing the complexities of the baseball audience experience.
Our insights into audience interest are derived from a detailed examination of recent data, specifically from the final 40 home games of each team in 2019, 2021, and 2022.
This research not only contributes to the field of sports analytics but also provides practical insights for baseball stakeholders seeking to optimize fan engagement strategies in the constantly evolving landscape of professional sports.

1 緒論

2 觀眾進場人數二階段模型
2.1 主場球隊勝率模型
2.2 觀眾進場人數模型

3 模型參數估計以及進場人數預測方法
3.1 廣義估計方程式 (Generalized Estimating Equations, GEE)
3.2 廣義線性模型 (Generalized Linear Model, GLM)
3.3 進場人數預測方法

4 2019, 2021, 2022 賽季美國職業棒球大聯盟進場人數分析
4.1 方法比較
4.2 主場球隊勝率分析結果
4.3 觀眾進場人數分析結果
4.4 各月份觀眾進場人數分析

5 總結

1. Liang, K. Y., \& Zeger, S. L. (1986). Longitudinal data analysis using generalized linear models. \textit{Biometrika}, 73(1), 13-22.

2. Tainsky, S., \& Winfree, J. A. (2010). Short-run demand and uncertainty of outcome in Major League Baseball. \textit{Review of Industrial Organization}, 37, 197-214.

3. Carroll, R. J., Wang, S., Simpson, D. G., Stromberg, A. J., \& Ruppert, D. (1998). The sandwich (robust covariance matrix) estimator. Unpublished manuscript.

4. McCullagh, P. (2019). Generalized linear models. Routledge.

5. Koning, R. H. (1999). The competitive balance in Dutch soccer.

6. Baade, R. A., \& Tiehen, L. J. (1990). An analysis of major league baseball attendance, 1969-1987. Journal of Sport and Social Issues, 14(1), 14-32.

7. Ahn, S. C., \& Lee, Y. H. (2014). Major League Baseball attendance: Long-term analysis using factor models. Journal of Sports Economics, 15(5), 451-477.

8. Jeong, J., \& Chung, S. (2001). Bootstrap tests for autocorrelation. Computational statistics \& data analysis, 38(1), 49-69.

9. Chan, N. H. (2011). Time series: Applications to finance with R and S-Plus (Vol. 837). John Wiley \& Sons.

10. Chai, T., \& Draxler, R. R. (2014). Root mean square error (RMSE) or mean absolute error (MAE). Geoscientific model development discussions, 7(1), 1525-1534.

11. Sokolova, M., \& Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information processing \& management, 45(4), 427-437.

12. Schruben, L. (1983). Confidence interval estimation using standardized time series. Operations Research, 31(6), 1090-1108.

13. Hardin, J. W., \& Hilbe, J. M. (2002). Generalized estimating equations. chapman and hall/CRC.

14. Hearst, M. A., Dumais, S. T., Osuna, E., Platt, J., \& Scholkopf, B. (1998). Support vector machines. IEEE Intelligent Systems and their applications, 13(4), 18-28.

15. Chen, T., \& Guestrin, C. (2016, August). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794).