透過您的圖書館登入
IP:216.73.216.100
  • 學位論文

使用機器學習演算法加入市場變數來預測財務危機

Using machine learning algorithms to predict financial distress by adding marketing variables

指導教授 : 石百達
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


過去文獻較多只用財務變數的組合來探討企業危機預測的機率,較少對於市場變數的組合有著墨,且考慮到全年度財報發佈時間在三月底前,若企業發生危機是在財報發布前,則沒有前一年度的財報可參考。本研究探討運用前兩年度財務變數再加上市場變數,是否能提高財務危機預測模型的準確率。 研究結果顯示,在決大多數的情況下,加入特定市場變數所訓練出的模型能有效提高預測力,且在眾多機器學習模型中,RF的預測能力最穩定,預測能力最準確。

並列摘要


Previous studies usually only use financial variables to establish financial distress forecasting models. However, if companies have financial crises before the financial reports are revealed, investors can’t use them to establish the models. This study will use the financial data of the previous two years and add market variables to build financial distress prediction models. The results show that adding marketing variables improve the performance of the models in the majority time. Compared to other machine learning algorithms, random forest is the best model in out-of-sample tests.

參考文獻


Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of Finance, 23, 589-609.
Beaver, W. H. (1966). Financial ratios as predictors of failure. Journal of Accounting Research, 4, 71-111.
Breiman, L. (1966). Random Forests. Machine Learning, 45, 5–32.
Bharath, S. T., & Shumway, T. (2008). Forecasting default with the Merton distance to default model. The Review of Financial Studies, 21, 1339-1369.
Erdogan, B. E. (2013). Prediction of bankruptcy using support vector machines: an application to bank bankruptcy. Journal of Statistical Computation and Simulation, 83, 1543-1555.

延伸閱讀