The 2007 financial crisis caused by the subprime mortgage crisis is the breakdown of trust that occurred between banks and the trigger for the global financial disturbances of 2008. Therefore, credit risk assessment/prediction is one of the most important management issues in the fields of accounting and finance. Even a fraction of improvement on a forecasting model’s performance, it will translate large amount of future savings.
Although many researches paid considerable attentions on prediction model construction by means of numerical ratio analysis, the work on managerial ability extraction from top managers that has been widely viewed as a main trigger for financial troubles is quite scarce. Apart from previous works on managerial ability extraction via conducting interviewing or questionnaires that full of subjectivity, this study follows the research done by Demerjian, Lev and McVay (2012) who performed mathematical linear programming approach to handle this task. Due to this method belongs data driven category, the outcome is objective. The numerical information collected form financial reports; managerial ability derived from linear programming technique inserted into support vector machine (SVM) which poses superior forecasting performance, better extreme value tolerance, and outstanding generalization ability, to construct the model for credit risk forecasting. The semiconductor industry plays an important role in global supply chain of electronic devices and laptops. This industry also received many financial incentives and government funding and it turns out to be an economic backbone of Taiwan’s economy. Thus, this study takes this specific industry as an example. The result indicates that the managerial ability plays an essential role in forecasting model construction and the introduced model is a promising alternative for credit risk forecasting. The managers can consider the outcomes derived from this study to adjust their business operation strategies and directions as well as formulate future policies to reach a goal of sustainable development.