Author(s)

YunShan Cai¹, MeiNa Li², HengYuan Fan^1*

Affiliation(s)

¹School of Statistics and Data Science, Southwestern University of Finance and Economics, Chengdu 611130, Sichuan, China.

²School of Management Science and Engineering, Southwestern University of Finance and Economics, Chengdu 611130, Sichuan, China.

Corresponding Author

HengYuan Fan

ABSTRACT

Olympic medal counts reflect both athletic strength and national soft power. Existing research often gives point estimates without confidence intervals, uses single models, and neglects factors like host‐country influence and coaching effects. To address these gaps, this study develops two complementary approaches: (1) an XGBoost regression model with Tree‐structured Parzen Estimator (TPE) optimization to predict gold, silver, and bronze medal counts (1988–2024 data) and construct confidence intervals from residuals; (2) a two‐way fixed effects Difference‐in‐Differences (DID) model to quantify the “great coach effect” by comparing China’s table tennis team before and after 2003 against control groups. The XGBoost model achieves R2 scores of 0. 842 for gold and 0. 850 for silver, providing credible intervals for 2028 predictions. The DID analysis shows elite coaches (e. g. , Liu Guoliang) increased China’s annual medal count by about three, with results robust under various specifications. These findings offer data‐driven guidance for National Olympic Committees in target setting, resource allocation, and coach investment, while presenting a generalized framework for evaluating talent effects in sports policy.

KEYWORDS

Olympic medal prediction; XGBoost; TPE optimization; Difference‐in‐Differences; Coach effect; Confidence interval

CITE THIS PAPER

YunShan Cai, MeiNa Li, HengYuan Fan. Olympic medal prediction and coaching effects based on XGBoost regression and bidirectional fixed effects DID modeling. World Journal of Information Technology. 2025, 3(3): 67-72. DOI: https://doi.org/10.61784/wjit3045.

REFERENCES

[1] Sayeed R, Hassan M T, Rahman M N, et al. Machine Learning Models for Predicting Olympic Medal Outcomes//2025 IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), Gwalior, India. IEEE, 2025, 3: 1-6. DOI: 10.1109/IATMSI64286.2025.10984687.

[2] Sagala N T M, Ibrahim M A. A Comparative Study of Different Boosting Algorithms for Predicting Olympic Medal//2022 IEEE 8th International Conference on Computing, Engineering and Design (ICCED), Sukabumi, Indonesia. IEEE, 2022: 1-4. DOI: 10.1109/ICCED56140.2022.10010351.

[3] Yang Y. Market Forecast using XGboost and Hyperparameters Optimized by TPE//2021 IEEE International Conference on Artificial Intelligence and Industrial Design (AIID), Guangzhou, China. IEEE, 2021: 7-10. DOI: 10.1109/AIID51893.2021.9456538.

[4] Zhao W, Sang S, Han S, et al. The Prediction of Coalbed Methane Layer in Multiple Coal Seam Groups Based on an Optimized XGBoost Model. Energies, 2024, 17(23): 6060.

[5] Zhao S, Cao J, Steve J. Research on Olympic medal prediction based on GA-BP and logistic regression model. F1000Research, 2025, 14: 245.

[6] Andrews D S, Meyer K E. How much does host country matter, really?. Journal of World Business, 2023, 58(2): 101413.

[7] Anchez-Fernandez P, Vaamonde-Liste A. Olympic medals: Success predictions for Río-2016. South African Journal for Research in Sport, Physical Education and Recreation, 2016, 38(3): 195-206.

[8] Nagpal P, Gupta K, Verma Y, et al. Paris Olympic (2024) Medal Tally Prediction//International Conference on Data Management, Analytics & Innovation. Singapore: Springer Nature Singapore, 2023, 662: 249-267. DOI: https://doi.org/10.1007/978-981-99-1414-2_20.

[9] Sayeed R, Hassan M T, Rahman M N, et al. Machine Learning Models for Predicting Olympic Medal Outcomes//2025 IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), Gwalior, India. IEEE, 2025, 3: 1-6. DOI: 10.1109/IATMSI64286.2025.10984687.

[10] Borusyak K, Jaravel X, Spiess J. Revisiting event-study designs: robust and efficient estimation. Review of Economic Studies, 2024, 91(6): 3253-3285.

[11] Young P, Jakeman A. Refined instrumental variable methods of recursive time-series analysis Part III. Extensions. International Journal of Control, 1980, 31(4): 741-764.

[12] Miller D L. An introductory guide to event study models. Journal of Economic Perspectives, 2023, 37(2): 203-230.

[13] Clarke D, Tapia-Schythe K. Implementing the panel event study. The Stata Journal, 2021, 21(4): 853-884.

[14] Hague C, McGuire C S, Chen J, et al. Coaches’ influence on team dynamics in sport: A scoping review. Sports Coaching Review, 2021, 10(2): 225-248.

[15] Gould D, Greenleaf C, Guinan D, et al. A survey of US Olympic coaches: Variables perceived to have influenced athlete performances and coach effectiveness. The sport psychologist, 2002, 16(3): 229-250.