Abstract
Capturing of relevant patterns in company's financial data and the implications on the reporting are important for various financial statement users to identify the triggers of the significant deficiencies and material weaknesses. The objective of this study is to construct a company-specific risk score for the companies' internal weaknesses, as well as to uncover the conditional relations between the independent predictors of firms' material weaknesses. To do so, Tree Augmented Naive Bayes (TAN) and Logistic Regression (LR) algorithms are employed to analyze the data obtained from COMPUSTAT (Research Insight) for one year before the Material Weakness in Internal Control (MWIC) disclosure on several operating and financial ratios such as total asset turnover, profitability, capital intensity, size, current ratio, and operating performance. The proposed TAN method provides novel information on the interactions among the predictors and the conditional probability of MWIC for a given set of relevant firm characteristics.
Original language | English |
---|---|
Pages | 971-976 |
Number of pages | 6 |
State | Published - 2018 |
Event | 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018 - Orlando, United States Duration: 19 May 2018 → 22 May 2018 |
Conference
Conference | 2018 Institute of Industrial and Systems Engineers Annual Conference and Expo, IISE 2018 |
---|---|
Country/Territory | United States |
City | Orlando |
Period | 19/05/18 → 22/05/18 |
Keywords
- Bayesian belief network
- Data mining
- Logistic regression
- Machine learning