Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/157786
|
| Title: | 應用機器學習對中國房地產上市公司違約預測之研究
Predicting Corporate Default in China’s Real Estate Sector : A Machine Learning Approach |
| Authors: | 張琳婕
Zhang, Lin-Jie |
| Contributors: | 李志宏
張琳婕
Zhang, Lin-Jie |
| Keywords: | 房地產違約風險
機器學習
違約預警系統
Real Estate Default Risk
Machine Learning
Default Early-Warning System |
| Date: | 2025 |
| Issue Date: | 2025-07-01 14:52:43 (UTC+8) |
| Abstract: | 中國房地產市場作為國民經濟支柱，近年來受政策調控、疫情衝擊與融資限制等多种因素影響，上市房地產企業違約事件頻發，對金融市場穩定構成潛在威脅。傳統違約預測方法多依賴單一財務指標，對系統性風險與非財務因素的整合預測能力有限。
本研究以CSMAR資料庫中2005年至2020年間中國房地產上市公司為樣本，結合中國統計年鑒中的宏觀經濟指標，構建涵蓋公司財務、償債能力與宏觀環境的多維預測變數。透過XGBoost、LightGBM與CatBoost等集成學習方法進行特徵篩選，最終確定18項關鍵指標。接著以隨機森林、支援向量機等模型進行比較分析，從準確率、召回率與AUC等多項指標評估模型性能，結果顯示隨機森林在整體預測表現上優於其他模型。並且該模型在2021-2023年的測試中成功預測實際違約的全部企業，與傳統財務模型Logistic相比，具有更高的適配度、準確性和較高的實用價值。
本研究不僅豐富了中國房地產違約風險預測領域的研究內容，也為政府監管部門、金融機構和投資者提供了一種可行的風險預警工具。未來研究可擴大樣本範圍，並引入更多非財務因素，以提高預測模型的全面性與準確性。
As a pillar of China’s national economy, the real estate sector has faced mounting pressure in recent years due to policy tightening, the impact of the COVID-19 pandemic, and financing restrictions. Consequently, default events among publicly listed real estate companies have occurred with increasing frequency, posing potential risks to the stability of financial markets. Traditional default prediction models largely rely on single financial indicators and are limited in their ability to integrate systemic risks and non-financial factors.
This study draws on data from the CSMAR database covering China’s listed real estate firms from 2005 to 2020, supplemented with macroeconomic indicators from the China Statistical Yearbook. It constructs a multidimensional set of predictive variables encompassing corporate financials, solvency indicators, and macroeconomic conditions. Using ensemble machine learning techniques such as XGBoost, LightGBM, and CatBoost for feature selection, 18 key predictors were identified. Multiple models, including Random Forest and Support Vector Machines, were evaluated based on accuracy, recall, and AUC. The Random Forest model demonstrated superior overall performance. Notably, it successfully predicted all actual default cases between 2021 and 2023, outperforming the traditional logistic regression model in terms of adaptability, accuracy, and practical value.
This research contributes to the growing body of literature on default risk prediction in China’s real estate sector and provides a viable early warning tool for regulators, financial institutions, and investors. Future studies could expand the sample scope and incorporate more non-financial variables to enhance the model’s comprehensiveness and predictive power. |
| Reference: | 英文參考文獻
Altman, E. I., Hu, X., & Yu, J. (2022). Has the Evergrande debt crisis rattled Chinese capital markets? A series of event studies and their implications. Finance Research Letters, 50, 103247. Elsevier BV.
Altman, E., Resti, A., & Sironi, A. (2004). Default Recovery Rates in Credit Risk Modelling: A Review of the Literature and Empirical Evidence. Economic Notes, 33(2), 183–208.
Anik, S. M. H., Gao, X., & Meng, N. (2023). Towards automated occupant profile creation in smart buildings: A machine learning-enabled approach for user persona generation. Energy and Buildings, 297, 113485.
Azizpour, S., Giesecke, K., & Schwenkler, G. (2018). Exploring the sources of default clustering. Journal of Financial Economics, 129(1), 154–183.
Barboza, F., Kimura, H., & Altman, E. (2017). Machine learning models and bankruptcy prediction. Expert Systems with Applications, 83, 405–417.
Bennett, R. L., Güntay, L., & Unal, H. (2015). Inside debt, bank default risk, and performance during the crisis. Journal of Financial Intermediation, 24(4), 487–513.
Bottazzi, G., Grazzi, M., Secchi, A., & Tamagni, F. (2011). Financial and economic determinants of firm default. Journal of Evolutionary Economics, 21(3), 373–406.
Chen, S. (2024). Analysis of China’s Real Estate Market Volatility and Financial Stability. Highlights in Business, Economics and Management, 28, 144–149.
Chen, Y., Cheung, K. C., Sun, R. Z., & Yam, S. C. P. (2024). A user guide of CART and random forests with applications in FinTech and InsurTech. Japanese Journal of Statistics and Data Science, 7(2), 999–1038.
Cincinelli, P., Pellini, E., & Urga, G. (2022). Systemic risk in the Chinese financial system: A panel Granger causality analysis. International Review of Financial Analysis, 82, 102179.
Cowden, C., Fabozzi, F. J., & Nazemi, A. (2019). Default Prediction of Commercial Real Estate Properties Using Machine Learning Techniques. Journal of Portfolio Management, 45(7), 55–67. Pageant Media.
Donovan, J., Jennings, J., Koharki, K., & Lee, J. (2021). Measuring credit risk using qualitative disclosure. Review of Accounting Studies, 26(2), 815–863.
Duca, J. V., Muellbauer, J., & Murphy, A. (2010). Housing markets and the financial crisis of 2007–2009: Lessons for the future. Journal of Financial Stability, 6(4), 203–217.
Episcopos, A., Pericli, A., & Hu, J. (1998). Commercial Mortgage Default: A Comparison of Logit with Radial Basis Function Networks. The Journal of Real Estate Finance and Economics, 17(2), 163–178.
Fischer, S. (1993). The role of macroeconomic factors in growth. Journal of Monetary Economics, 32(3), 485–512.
Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, 29(5), 1189–1232. Institute of Mathematical Statistics.
Fung, H.-G., Jeng, J.-L., & Liu, Q. “Wilson.” (2010). Development of China’s Real Estate Market. The Chinese Economy, 43(1), 71–92. Routledge.
Han, S., & Zhou, X. (2014). Informed Bond Trading, Corporate Yield Spreads, and Corporate Default Prediction. Management Science, 60(3), 675–694. INFORMS.
Hancock, J. T., & Khoshgoftaar, T. M. (2020). CatBoost for big data: An interdisciplinary review. Journal of Big Data, 7(1), 94.
Huang, B., Yao, X., Luo, Y., & Li, J. (2023). Improving financial distress prediction using textual sentiment of annual reports. Annals of Operations Research, 330(1), 457–484.
Huang, T. (2023). Working Paper 23-5: Why China’s housing policies have failed.
Huang, Y., Wang, Z., & Jiang, C. (2024). Diagnosis with incomplete multi-view data: A variational deep financial distress prediction method. Technological Forecasting and Social Change, 201, 123269.
Islam, M. S., Alam, M. S., Bin Hasan, S., & Mollah, S. (2022). Firm-level political risk and distance-to-default. Journal of Financial Stability, 63, 101082.
Kallberg, J. G., Liu, C. H., & Pasquariello, P. (2014). On the Price Comovement of U.S. Residential Real Estate Markets. Real Estate Economics, 42(1), 71–108.
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T. Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30.
Li, S., Liu, J., Dong, J., & Li, X. (2021). 20 Years of Research on Real Estate Bubbles, Risk and Exuberance: A Bibliometric Analysis. Sustainability, 13(17), 9657. Multidisciplinary Digital Publishing Institute.
Lu, L., & Keller, A. (2022a). Is it China’s Lehman Brothers moment? Unveiling Evergrande debt crisis, financial risks, and regulatory implications. Law and Financial Markets Review, 16(1–2), 133–144. Routledge.
Lu, L., & Keller, A. (2022b). Is it China’s Lehman Brothers moment? Unveiling Evergrande debt crisis, financial risks, and regulatory implications. Law and Financial Markets Review, 16(1–2), 133–144. Routledge.
Ma, X., Che, T., & Jiang, Q. (2025). A three-stage prediction model for firm default risk: An integration of text sentiment analysis. Omega, 131, 103207.
Ma, Y., Zhang, P., Duan, S., & Zhang, T. (2023). Credit default prediction of Chinese real estate listed companies based on explainable machine learning. Finance Research Letters, 58, 104305.
Mostafaei, K., Maleki, S., Zamani Ahmad Mahmoudi, M., & Knez, D. (2022). Risk management prediction of mining and industrial projects by support vector machine. Resources Policy, 78, 102819.
Mselmi, N., Lahiani, A., & Hamza, T. (2017). Financial distress prediction: The case of French small and medium-sized firms. International Review of Financial Analysis, 50, 67–80.
Nazemi, A., & Fabozzi, F. J. (2024). Interpretable machine learning for creditor recovery rates. Journal of Banking & Finance, 164, 107187.
Patel, K., & Vlamis, P. (2006). An Empirical Estimation of Default Risk of the UK Real Estate Companies. The Journal of Real Estate Finance and Economics, 32(1), 21–40.
Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2018). CatBoost: unbiased boosting with categorical features. Advances in neural information processing systems, 31.
Qian, H., Wang, B., Yuan, M., Gao, S., & Song, Y. (2022). Financial distress prediction using a corrected feature selection measure and gradient boosted decision tree. Expert Systems with Applications, 190, 116202.
Radovanovic, J., & Haas, C. (2023). The evaluation of bankruptcy prediction models based on socio-economic costs. Expert Systems with Applications, 227, 120275.
Tang, P., Tang, T., & Lu, C. (2024). Predicting systemic financial risk with interpretable machine learning. The North American Journal of Economics and Finance, 71, 102088.
Thilini, M., & Wickramaarachchi, N. C. (2019). Risk assessment in commercial real estate development: An application of analytic network process. Journal of Property Investment & Finance, 37(5), 427–444.
Vlamis, P. (2007). Default Risk of the UK Real Estate Companies: Is There a Macro-economy Effect? The Journal of Economic Asymmetries, 4(2), 99–117.
Zhang, J. (2024). Impact of an improved random forest-based financial management model on the effectiveness of corporate sustainability decisions. Systems and Soft Computing, 6, 200102.
Zhao, C., & Liu, F. (2023). Impact of housing policies on the real estate market—Systematic literature review. Heliyon, 9(10), e20704.
Zurek, M. (2022). Real Estate Markets and Lending: Does Local Growth Fuel Risk? Journal of Financial Services Research, 62(1), 27–59.
中文參考文獻
刘洪波、刘俊莹（2023），我国房地产企业的信用风险评价研究，征信，41(3)，66–72。
史富莲、石亚玲（2007），Z值模型在房地产业上市公司财务预警分析中的应用，会计之友，(2)，91–92。
周曲文（2024），基于机器学习技术的房地产企业债券违约风险预警，微型电脑应用，40(8)，207-210，215。
姚鹏（2019），基于KMV模型的中国上市房地产公司信用风险评估研究，中国资产评估，(6)，11–19。
尹剑、高杰林、张小成 (2023)，基于中国信用债市场的房地产企业信用风险评价研究，黑龙江金融，(5)，80–86。
张红、李洋、黄硕（2013），中国房地产上市公司财务危机预警研究，河南社会科学，21(3)，76–79。
彭宇超、张搏（2024），我国房地产风险成因分析及未来发展展望，中国房地产金融， (5)，37–43。
李建中、武铁梅（2010），基于因子—Logistic模型的房地产业上市公司财务预警分析，哈尔滨商业大学学报（社会科学版），(5)，89-93，104。
王俊籽、刘澜涛（2017），基于Logistic模型的我国商业银行房地产信贷风险研究. 经济与管理评论，(2)，86–95。
王磊、蒋建旺、王冀宁、陈庭强（2024），房地产企业流动性风险的成因与治理路径研究，财会通讯，(16)，125-130，149。
胡胜、雷欢欢、胡华强（2018），基于Logistic模型的我国房地产企业信用风险度量研究，中国软科学，(12)，157–164。
袁东（2023），“三道红线”政策对房地产企业信用风险的影响，哈尔滨学院学报， 44(12)，35。
郑晓云、李建华（2015），房地产上市公司财务预警实证研究，会计之友，(9)，72–76。 |
| Description: | 碩士
國立政治大學
財務管理學系
112357041 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0112357041 |
| Data Type: | thesis |
| Appears in Collections: | [財務管理學系] 學位論文
|
Files in This Item:
| File |
Size | Format | |
|---|
| 704101.pdf | 913Kb | Adobe PDF | 0 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
