Forward-looking: A machine learning approach in predicting corporate delisting in the Philippine Stock Exchange

• Main Authors: Abrero, Jan Paul Ringo S., Beltran, Alexander Christoph P., Limsui, Rainne Chelsea O., Yeung, Kharl Christian T.
• Format: text
• Language: English
• Published: Animo Repository 2017
• Subjects: Business enterprises > Philippines > Finance; Business enterprises > Philippines; Finance and Financial Management
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/6464
• Institution: De La Salle University

Delisting is the removal of a listed security from the exchange where it is traded. Its drawbacks extend to the liquidity status of the shareholder, access to funding sources, and immediate assessment of the enterprise, which could possibly result in financial loss and consumer confidence reduction. With this, the study explores the predicting capabilities of Altman Z-score model, logistic regression analysis and artificial neural network on the probability of corporate delisting in the Philippine Stock Exchange (PSE). With a sample of twenty-six (26) delisted corporations and seventy-eight (78) publicly-listed corporations from 1995-2016, the researchers employed a machine learning approach to identify the most effective predictive model. Using T-test and chi-squared test, results showed that quick ratio (QUICK), total debt to equity ratio (DEBTEQ), and degree of financial leverage (DFL) were statistically significant and listing status was only statistically dependent to minimum public ownership compliance (MPO). Empirical results showed that artificial neural network was the most effective model with 77.42% accuracy and 67.00% precision using whole data set one (1) prior to delisting and 70.97% accuracy and 56.67% precision using whole data set two (2) years prior to delisting.