Analyzing initial public offerings' short-term performance using decision trees and SVMs

• Underpricing of stocks in IPOs is an important and challenging research topic.
• Decision tree models are shown to be powerful tools to analyze early IPO dynamics.
• C5.0 algorithm produced significantly better prediction results than CART.
• An information fusion-based sensitivity analysis was used to identify variable importance.
• Market sentiment, sales and turnover rates were identified as the most important variables.

In this study, we investigated underpricing of Turkish companies in the initial public offerings (IPOs) issued and traded on Borsa Istanbul between 2005 and 2013. The underpricing of stocks in IPOs, or essentially leaving money on the table, is considered as an important, challenging and worthy research topic in literature. Within the proposed framework, the IPO performance in the short run and the factors that affect this short run performance were analyzed. Popular machine learning methods – several decision tree models and support vector machines – were developed to investigate the major factors affecting the short-term performance of initial IPOs. A k-fold cross validation methodology was used to assess and contrast the performance of the predictive models. An information fusion-based sensitivity analysis was performed to combine the values of individual variable importance results into a common representation. The results showed that there was underpricing in the initial public offerings of Turkish companies, although it was not as high as the underpricing determined in developed markets. The market sentiment, the annual sales amounts, the total assets turnover rates, IPO stocks sales methods, the underwriting methods, the offer prices, debt ratio, and number of shares sold were among the most influential factors affecting the short term performance of initial public offerings of Turkish companies.