Loading...

Bankruptcy Forecasting for Companies and Providing Counterfactual Scenarios to Change the Bankruptcy Class According to Financial Statement Data

Haji Hajikolaei, Maryam | 2023

35 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56575 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Akhavan Niaki, Taghi
  7. Abstract:
  8. Bankruptcy is an important issue in the economy that can have extensive financial and social consequences on individuals and society. Timely warning to managers and providing analysis may prevent bankruptcy. Many studies have been conducted on the application and implementation of machine learning techniques to predict bankruptcy. Many bankruptcy prediction models produce incomprehensible outputs for the user. Therefore, they are called black box algorithms. Implementation of advanced models inevitably requires interpretability for users to understand the result and trust. Since most machine learning methods are "black box", explainable AI, which aims to provide explanations to users, has become an important research topic. This thesis is an attempt to provide solutions to financial managers so that they can predict their company’s bankruptcy in the future by using financial ratios and in case of possible bankruptcy by receiving counterfactual scenarios, gain the necessary insight to make appropriate decisions to avoid bankruptcy. Therefore, this research includes a machine learning model that predicts the probability of bankruptcy of companies with high accuracy, and further, by using algorithms to generate counterfactual explanations, it proposes solutions to change the company's situation. The validity of the proposed model has been evaluated with the help of the dataset of Polish companies taken from the EMSI database, as a case study, from three aspects: The first evaluation shows that the proposed machine learning model performs better than the current researches in the literature. The second evaluation, by examining the indicators, shows that the models for producing counterfactual explanations are in accordance with the assumptions of the financial literature and are applicable in reality. Finally, the third evaluation with a survey of financial managers shows that the proposed scenarios lead to the creation of insights to make decisions about changing financial policies and preventing bankruptcy
  9. Keywords:
  10. Machine Learning ; Bankruptcy ; Explainable Artificial Intelligence ; Counterfactual Explanations ; Black Box Model

 Digital Object List

 Bookmark

No TOC