Saul Kornik

Machine Learning for Corporate Failure Prediction: an Empirical Study of South African Companies Saul Kornik

Corporate failure is an essential component of an efficient market economy. It allows for the recycling of financial, human and physical resources into more productive organisations. However, many stakeholders, have an interest in the financial health of a firm, as the failure of the corporation can have a significant impact on the costs to all parties. Machine learning can broadly be defined as the field of study that concentrates on algorithms that have the ability to learn. This is in direct contrast to expert systems that are automated with a set of predetermined rules for the classification of the independent variable. Machine learning techniques are adept at finding potential solutions to highly complex problems. In this research, support vector machines and genetic algorithms were applied to the problem of corporate failure prediction (a complex, non-linear problem) with great effect. The book ends by showing, mathematically, the relationship between support vector machines and kernel ridge regression.