本研究針對 2020-2023 年美國股票市場，科技產業中智慧機器人相關產品或技術發布事件對競爭者公司股價的影響。本研究旨在探討智慧機器人技術發布後如何改變投資者對相關競爭者的市場評估，新產品發布不僅影響發布企業本身，也會影響到同業公司在市場中的相對競爭力和未來收益預期。競爭者為與事件公司在同一產業內，提供相似技術或產品的主要廠商。本研究以事件研究法為基礎，選定多個智慧機器人重大產品發布為分析樣本，定義事件窗口為發布日前後 3 天。通過計算異常報酬（Abnormal Returns, AR），分析事件對競爭者股價的短期影響，並應用機器學習分類器（包括 eXtreme Gradient Boosting、支持向量迴歸和隨機森林）進行預測。透過兩種資料抽取方式（隨機抽取資料和隨機抽取公司），比較各種模型的表現。
迴歸模型結果方面，隨機抽取資料的模型擬合度較高；隨機抽取事件的模型擬合度稍低。在機器學習模型結果中，隨機森林模型在 MSE 與 𝑅2 方面表現最佳，SVR 模型在 1% 命中率上表現最佳。
綜合結論顯示，隨機抽取資料的迴歸模型擬合度較高，解釋變數能力更強，而隨機抽取事件的機器學習模型預測誤差較小，表現更穩定。以上兩種方式在高精度預測上的能力非常接近，均能提供較好的預測結果。

This study focuses on the impact of product or technology release events related to intelligent robots in the technology industry on the stock prices of competitor companies in the U.S. stock market from 2020 to 2023. The research aims to explore how the release of intelligent robot technologies influences investors’ market evaluations of related competitors. The release of new products not only affects the company making the announcement but also impacts the relative competitiveness and future earnings expectations of peer companies in the market.Competitors are defined as major firms within the same industry as the event company that provide similar technologies or products.This study is based on the event study methodology, selecting multiple major intelligent robot product releases as analytical samples and defining the event window as three days before and after the release date. By calculating abnormal returns (AR), the shortterm impact of events on competitors’ stock prices is analyzed. Additionally, machine learning classifiers, including eXtreme Gradient Boosting (XGBoost), support vector regression (SVR), and random forests, are applied for prediction. Two data extraction methods(random data sampling and random company sampling)are used to compare the performance of various models.

論文審定書.................................................................... i
誌謝...........................................................................ii
摘要..........................................................................iii
Abstract......................................................................iv
第一章介紹.................................................................1
1.1研究背景................................................................1
1.2研究動機................................................................2
1.3研究流程與架構..........................................................2
第二章文獻探討............................................................5
2.1事件研究法..............................................................5
2.2競爭者分析..............................................................5
2.3機器學習................................................................7
第三章資料及研究方法....................................................10
3.1事件研究法.............................................................10
3.2資料集..................................................................11
3.3迴歸分析...............................................................14
3.4機器學習方法...........................................................16
3.4.1極端梯度提升法(eXtremeGradientBoosting) .....................16
3.4.2支持向量迴歸(SVR)..............................................20
3.4.3隨機森林（RandomForest）.....................................22
3.5判斷模型指標...........................................................25
3.5.1均方誤差(MSE)..................................................25
3.5.2決定係數(𝑅2) ....................................................26
3.5.3命中率(HitRate).................................................28
第四章實驗結果...........................................................30
4.1模型重要變數...........................................................30
4.2事件前與事件後重要變數比較...........................................33
4.3隨機抽取資料與隨機抽取公司...........................................38
4.4兩種抽取方式結果比較..................................................42
第五章結論與未來展望....................................................45
5.1研究結論...............................................................45
5.2未來展望...............................................................45
參考文獻.....................................................................47

[1] Adom,A. Y., Nyarko, I. K., & Som, G. N.K. (2016). Competitor analysis in strategic management: Is it a worthwhile managerial practice in contemporary times. Journal of Resources Development and Management, 24(1), 116–127.
[2] Al-Ghuribi, S. M., & Noah, S. A. M. (2019). Multi-criteria review-based recommender system–the state of the art. IEEE Access, 7, 169446–169468.
[3] Alsini, A., Huynh, D. Q., & Datta, A. (2020). Hit ratio: An evaluation metric for hashtag recommendation. arXiv preprint arXiv:2010.01258.
[4] Alsoboa, S. S., & Alalaya, M. M. (2015). Practices of competitor accounting and its influence on the competitive advantages: An empirical study in jordanian manufacturing companies. Global Journal of Management and Business Research, 15(3),12–23.
[5] Asselman, A., Khaldi, M.,&Aammou,S. (2023). Enhancing theprediction of student performance based on the machine learning xgboost algorithm. Interactive Learning Environments, 31(6), 3360–3379.
[6] Ball, R., Gerakos, J., Linnainmaa, J. T., & Nikolaev, V. (2020). Earnings, retained earnings, and book-to-market in the cross section of expected returns. Journal of Financial Economics, 135(1), 231–254.
[7] Banachewicz, K., & Massaron, L. (2022). The kaggle book: Data analysis and machine learning for competitive data science. Packt Publishing Ltd.
[8] Banz, R. W. (1981). The relationship between return and market value of common stocks. Journal of financial economics, 9(1), 3–18.
[9] Beaver, W. H., McNichols, M. F., & Wang, Z. Z. (2020). Increased market response to earnings announcements in the 21st century: An empirical investigation. Journal of Accounting and Economics, 69(1), 101244.
[10] Beck, T., Demirguc-Kunt, A., & Levine, R. (2005). Smes, growth, and poverty: Cross-country evidence. Journal of economic growth, 10, 199–229.
[11] Behnken, D. W., & Draper, N. R. (1972). Residuals and their variance patterns.Technometrics, 14(1), 101–111.
[12] Bentéjac, C., Csörgő, A., & Martínez-Muñoz, G. (2021). A comparative analysis of gradient boosting algorithms. Artificial Intelligence Review, 54, 1937–1967.
[13] Bhandari, A. (2020). Key difference between r-squared and adjusted r-squared for regression analysis. Analytics Vidhya.
[14] Biau, G. (2012). Analysis of a random forests model. The Journal of Machine Learning Research, 13(1), 1063–1095.
[15] Blum, A., Lafferty, J., Rwebangira, M. R., & Reddy, R. (2004). Semi-supervised learning using randomized mincuts. In Proceedings of the twenty-first international conference on machine learning (p. 13).
[16] Botchkarev, A. (2018). Performance metrics (error measures) in machine learning regression, forecasting and prognostics: Properties and typology. arXiv preprint arXiv:1809.03006.
[17] Breiman, L. (2001). Random forests. Machine learning, 45, 5–32.
[18] Brigham, E. F., & Houston, J. F. (2013). Fundamentals of financial management.South-Western Cengage Learning.
[19] Brown,S.J.,&Warner, J.B. (1980). Measuring security price performance. Journal of financial economics, 8(3), 205–258.
[20] Brown, S. J., & Warner, J. B. (1985). Using daily stock returns: The case of event studies. Journal of financial economics, 14(1), 3–31.
[21] Bukowski, M., Kurek, J., Świderski, B., & Jegorowa, A. (2024). Custom loss functions in xgboost algorithm for enhanced critical error mitigation in drill-wear analysis of melamine-faced chipboard. Sensors, 24(4), 1092.
[22] Cao, Y. (2022). Application of machine learning algorithms in financial market risk prediction. In The 2021 international conference on machine learning and big data analytics for iot security and privacy: Spiot-2021 volume 1 (pp. 161–168).
[23] Castro-Iragorri, C. (2019). Does the market model provide a good counterfactual for event studies in finance? Financial Markets and Portfolio Management, 33, 71–91.
[24] Chen. (2016). Story and lessons behind the evolution of xgboost.
[25] Chen, J., Akash, A. K., & Suzuki, Y. (2021). Explore optimal degree of parallelism for distributed xgboost training (Tech. Rep.).
[26] Chen, M., Mao, S., & Liu, Y. (2012). Data mining for the internet of things: Literature review and challenges. Journal of Network and Computer Applications, 36(2),244–257.
[27] Chicco, D., & Jurman, G. (2020). The advantages of the matthews correlation coefficient (mcc) over f1 score and accuracy in binary classification evaluation. BMC genomics, 21, 1–13.
[28] Corrado, C.J. (2011). Event studies: Amethodology review. Accounting&Finance,51(1), 207–234.
[29] Criminisi, A., Shotton, J., & Konukoglu, E. (2011). Decision forests for classification, regression, density estimation, manifold learning and semi-supervised learning.Microsoft Research Cambridge, Tech. Rep. MSRTR-2011-114, 5(6), 12.
[30] Cutler, A., Cutler, D.R.,&Stevens, J.R. (2012). Random forests. Ensemble machine learning: Methods and applications, 157–175.
[31] Cygler, J., & Sroka, W. (2017). Coopetition disadvantages: The case of the high tech companies. Engineering economics, 28(5), 494–504.
[32] Deng,L.,&Liu,Y. (2018). Deep learning in natural language processing. Springer.
[33] Deng, L., & Platt, J. (2014). Ensemble deep learning for speech recognition. In Proc. interspeech.
[34] Dereli, D. D. (2015). Innovation management in global competition and competitive advantage. Procedia-Social and behavioral sciences, 195, 1365–1370.
[35] Devore, J. L. (2000). Probability and statistics. Pacific Grove: Brooks/Cole.
[36] Dilshad, M. N. (2013). Profitability analysis of mergers and acquisitions: an event study approach. Business and economic research, 3(1).
[37] Eisenhardt, K. M., & Zbaracki, M. J. (2000). Strategy as strategic decision making.Sloan Management Review, 43(2), 13–20.
[38] El Ghoul, S., Guedhami, O., Mansi, S. A., & Sy, O. (2023). Event studies in international finance research. Journal of International Business Studies, 54(2), 344.
[39] Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun,S. (2017). Dermatologist-level classification of skin cancer with deep neural networks.Nature, 542(7639), 115–118.
[40] Evans, D. S. (1987). Tests of alternative theories of firm growth. journal of political economy, 95(4), 657–674.
[41] Fama, E. F. (1970). Efficient capital markets. Journal of finance, 25(2), 383–417.
[42] Fama, E. F., Fisher, L., Jensen, M. C., & Roll, R. (1969). The adjustment of stock prices to new information. International economic review, 10(1), 1–21.
[43] Fama, E. F., & French, K. R. (1992). The cross-section of expected stock returns. the Journal of Finance, 47(2), 427–465.
[44] Fong, S. (2012). Framework of competitor analysis by monitoring information on the web. Journal of emerging technologies in web intelligence, 4(1), 77–83.
[45] Fuchs, C., & Diamantopoulos, A. (2010). Evaluating the effectiveness of brand positioning strategies from a consumer perspective. European Journal of Marketing,44(11/12), 1763–1786.
[46] Getz, D., & Page, S. J. (2019). Event studies: Theory, research and policy for planned events. Routledge.
[47] Ghanem, D., & Rosvall, D. (2014). Major world events impact on stock market prices:-an event study.
[48] Ghodrati, A., & Zulkifli, N. (2013). The impact of 5s implementation on industrial organizations’performance. International journal of business and management invention, 2(3), 43–49.
[49] Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press.
[50] Guo, L., Sharma, R., Yin, L., Lu, R., & Rong, K. (2017). Automated competitor analysis using big data analytics: Evidence from the fitness mobile app business.Business Process Management Journal, 23(3), 735–762.
[51] Guo, M., Kuai, Y., & Liu, X. (2020). Stock market response to environmental policies: Evidence from heavily polluting firms in china. Economic Modelling, 86,306–316.
[52] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the ieee conference on computer vision and pattern recognition (pp. 770–778).
[53] Hirschman, A. O. (1964). The paternity of an index. American economic review, 54(5).
[54] Hodson, T. O. (2022). Root mean square error (rmse) or mean absolute error (mae):When to use them or not. Geoscientific Model Development Discussions, 2022, 1–10.
[55] Jensen, M. C. (1978). Some anomalous evidence regarding market efficiency. Journal of financial economics, 6(2/3), 95–101.
[56] Kaiser, L., Babaeizadeh, M., Milos, P., Osinski, B., Campbell, R. H., Czechowski,K., … others (2019). Model-based reinforcement learning for atari. arXiv preprint arXiv:1903.00374.
[57] Kamilaris, A., & Prenafeta-Boldú, F. X. (2018). Deep learning in agriculture: A survey. Computers and Electronics in Agriculture, 147, 70–90.
[58] Karch, J. (2020). Improving on adjusted r-squared. Collabra: Psychology, 6(1).
[59] Konchitchki, Y., &O’Leary, D.E. (2011). Event study methodologies in information systems research. International Journal of Accounting Information Systems, 12(2), 99 -115.
[60] Koschate-Fischer, N., & Schandelmeier, S. (2014). A guideline for designing experimental studies in marketing research and a critical discussion of selected problem areas. Journal of Business Economics, 84, 793–826.
[61] Kramer,O.,etal. (2013). Dimensionality reduction with unsupervised nearest neighbors (Vol. 51). Springer.
[62] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25.
[63] Kumar,S.,Sharma,D.,Rao,S.,Lim,M.,&Mangla,S. (2021). Artificial intelligence and machine learning in finance: Identifying foundations, themes, and research clusters from bibliometric analysis. Annals of Operations Research.
[64] Leiva, R.G., Anta, A.F., Mancuso, V.,&Casari, P. (2019). A novel hyperparameter-free approach to decision tree construction that avoids overfitting by design. Ieee Access, 7, 99978–99987.
[65] Leonidou, L. C., Leonidou, C. N., Fotiadis, T. A., & Zeriti, A. (2013). Resources and capabilities as drivers of hotel environmental marketing strategy: Implications for competitive advantage and performance. Tourism management, 35, 94–110.
[66] Levine, S. S., Bernard, M., & Nagel, R. (2017). Strategic intelligence: The cognitive capability to anticipate competitor behavior. Strategic Management Journal, 38(12), 2390–2423.
[67] Li, X., Yang, P., & Jin, Z. (2017). Applications of machine learning in credit risk assessment: A survey. Journal of Risk and Financial Management, 10(2), 17.
[68] Li, Y., Liu, H., Xu, X., Xu, Y., Zhang, X., & Dong, X. (2019). Applications of machine learning in plant pathology. Agriculture, 9(8), 125.
[69] Liargovas, P., & Repousis, S. (2011). The impact of mergers and acquisitions on the performance of the greek banking sector: An event study approach. International Journal of Economics and Finance, 3(2), 89–100.
[70] Linkins, L.-A., Dans, A. L., Moores, L. K., Bona, R., Davidson, B. L., Schulman, S.,&Crowther, M. (2012). Treatment and prevention of heparin-induced thrombocytopenia: antithrombotic therapy and prevention of thrombosis: American college of chest physicians evidence-based clinical practice guidelines. Chest, 141(2), e495S–e530S.
[71] Liu, L., & Chen, Q. (2020). How to compare market efficiency? the sharpe ratio based on the arma-garch forecast. Financial Innovation, 6(1), 38.
[72] Liu, Y., & Liang, L. (2015). Evaluating and developing resource-based operations strategy for competitive advantage: an exploratory study of finnish high-tech manufacturing industries. International Journal of Production Research, 53(4), 1019–1037.
[73] Loke, W. H. (1989). Diagnostic evaluations using signal detection analysis. Indian Journal of Psychological Medicine, 12(1), 87–91.
[74] Lopez de Prado, M. (2018). Advances in financial machine learning. John Wiley & Sons.
[75] Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, 30.
[76] MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of economic literature, 35(1), 13–39.
[77] Martinez-Blasco, M., Serrano, V., Prior, F., & Cuadros, J. (2023). Analysis of an event study using the fama–french five-factor model: teaching approaches including spreadsheets and the r programming language. Financial Innovation, 9(1), 76.
[78] Miller, D. L. (2023). An introductory guide to event study models. Journal of Economic Perspectives, 37(2), 203–230.
[79] Mitchell, R., & Frank, E. (2017). Accelerating the xgboost algorithm using gpu computing. PeerJ Computer Science, 3, e127.
[80] Modigliani, F., & Miller, M. H. (1958). The cost of capital, corporation finance and the theory of investment. The American economic review, 48(3), 261–297.
[81] Murphy, K. P. (2012). Machine learning: a probabilistic perspective. MIT press.
[82] Nazareth, N., & Reddy, Y. (2023). Financial applications of machine learning: A literature review. Expert Systems with Applications, 219, 119640.
[83] Nozari, H., Sadeghi, M.E.,etal. (2021). Artificial intelligence and machine learning for real-world problems(asurvey). International Journal of InnovationinEngineering,1(3), 38–47.
[84] Palepu, K.G., Healy, P.M.,Wright,S., Bradbury, M.,&Coulton, J. (2020). Business analysis and valuation: Using financial statements. Cengage AU.
[85] Palia, A., & De Ryck, J. (2015). Assessing competitor strategic business units with the competitor analysis package. In Developments in business simulation and experiential learning: Proceedings of the annual absel conference (Vol. 42).
[86] Pate, A., Riley, R. D., Collins, G. S., Van Smeden, M., Van Calster, B., Ensor, J., & Martin, G. P. (2023). Minimum sample size for developing a multivariable prediction model using multinomial logistic regression. Statistical Methods in Medical Research,32(3), 555–571.
[87] Peng, Y. (2022). A survey on modern recommendation system based on big data. arXiv preprint arXiv:2206.02631.
[88] Porter, M. E. (1979). The structure within industries and companies’ performance.The review of economics and statistics, 214–227.
[89] Porter, M. E. (1980). Competitive strategy: Techniques for analyzing industries and competitors. Free Press.
[90] Porter, M. E. (2008). The five competitive forces that shape strategy. Harvard business review, 86(1), 78.
[91] Ricci, F., Rokach, L., & Shapira, B. (2011). Introduction to recommender systems handbook. Springer.
[92] Rua, O. L., & Santos, C. (2022). Linking brand and competitive advantage: The mediating effect of positioning and market orientation. European research on manage ment and business economics, 28(2), 100194.
[93] Sandler, D. H., &Sandler, R. (2014). Multiple event studies in public finance and labor economics: A simulation study with applications. Journal of Economic and Social Measurement, 39(1-2), 31–57.
[94] Satyro, W. C., Sacomano, J. B., Contador, J. C., Almeida, C. M., & Giannetti, B. F.(2017). Process of strategy formulation for sustainable environmental development:Basic model. Journal of Cleaner Production, 166, 1295–1304.
[95] Shalev-Shwartz, S., & Ben-David, S. (2014). Understanding machine learning: From theory to algorithms. Cambridge university press.
[96] Simons, R. (2019). The role of management control systems in creating competitiveadvantage: new perspectives. In Management control theory (pp. 173–194). Routledge.
[97] Singh, B., Kumar, R., & Singh, V. P. (2022). Reinforcement learning in robotic applications: a comprehensive survey. Artificial Intelligence Review, 55(2), 945–990.
[98] Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations. W.Strahan and T. Cadell.
[99] Srinivas, P., & Katarya, R. (2022). hyoptxg: Optuna hyper-parameter optimization framework for predicting cardiovascular disease using xgboost. Biomedical Signal Processing and Control, 73, 103456.
[100] Stamate,D.,Alghamdi,W.,Stahl,D.,Logofatu, D.,&Zamyatin,A. (2018). Pidt: A novel decision tree algorithm based on parameterised impurities and statistical pruning approaches. In Artificial intelligence applications and innovations: 14th ifip wg 12.5 international conference, aiai 2018, rhodes, greece, may 25–27, 2018, proceedings 14 (pp. 273–284).
[101] Steck, H. (2010). Training and testing of recommender systems on data missing not at random. In Proceedings of the 16th acm sigkdd international conference on knowledge discovery and data mining (pp. 713–722).
[102] Steel, R. G. D., Torrie, J. H., et al. (1960). Principles and procedures of statistics.Principles and procedures of statistics..
[103] Sudhaman, K., Akuthota, M., & Chaurasiya, S. K. (2022). A review on the different regression analysis in supervised learning. Bayesian Reasoning and Gaussian Processes for Machine Learning Applications, 15–32.
[104] Sun, W., Su, F., & Wang, L. (2018). Improving deep neural networks with multilayer maxout networks and a novel initialization method. Neurocomputing, 278, 34-40.
[105] Sutton, R. S., & Barto, A. G. (2018). Reinforcement learning: An introduction.MIT press.
[106] Tharwat, A. (2021). Classification assessment methods. Applied computing and informatics, 17(1), 168–192.
[107] Topol, E. (2019). Deep medicine: How artificial intelligence can make healthcare human again. Basic Books.
[108] Upson, J. W., Ketchen Jr, D. J., Connelly, B. L., & Ranft, A. L. (2012). Competitor analysis and foothold moves. Academy of Management Journal, 55(1), 93–110.
[109] Vapnik, V. (2013). The nature of statistical learning theory. Springer science & business media.
[110] Wang, H., Wang, G., Li, G., Peng, J., & Liu, Y. (2016). Deep belief network based deterministic and probabilistic wind speed forecasting approach. Applied energy, 182,80–93.
[111] Wang, M., Sha, Z., Huang, Y., Contractor, N., Fu, Y., & Chen, W. (2016). Forecasting technological impacts on customers’co-consideration behaviors: a data-driven networkanalysis approach. In International design engineering technical conferences and computers andinformation in engineering conference (Vol. 50107, p. V02AT03A040).
[112] Weinberger, K.Q.,&Saul,L.K. (2006). Unsupervised learning of image manifolds by semidefinite programming. International journal of computer vision, 70, 77–90.
[113] West, D. C., Ford, J. B., & Ibrahim, E. (2015). Strategic marketing: creating competitive advantage. Oxford University Press, USA.
[114] Yang, W., & Meyer, K. E. (2015). Competitive dynamics in an emerging economy: Competitive pressures, resources, and the speed of action. Journal of Business Research, 68(6), 1176–1185.
[115] Zhang, D., & Gong, Y. (2020). The comparison of lightgbm and xgboost coupling factor analysis and prediagnosis of acute liver failure. Ieee Access, 8, 220990–221003.
[116] Zhang, Y., Zhang, X., Razbek, J., Li, D., Xia, W., Bao, L., … Cao, M. (2022).Opening the black box: interpretable machine learning for predictor finding of metabolic syndrome. BMC Endocrine Disorders, 22(1), 214.
[117] Zheng,H., Yuan,J.,&Chen,L. (2017). Short-term load forecasting using emd-lstmneural networks with a xgboost algorithm for feature importance evaluation. Energies, 10(8), 1168.