Identifying poorly performing listed firms using data analytics

This study presents a teaching case that analyzes the applicability of Z-Score bankruptcy prediction model to manufacturing firms listed in Hong Kong. Although has been studied extensively, there are very few studies context Kong stock market. Given market high retail investor participation and low liquidity, whether is relevant investors an important but unanswered question. The predicts by considering financial ratios involving firm solvency, profitability, leverage, activity. Financial return data on Stock Exchange from 1981 2020 collected Thomson Reuters Datastream examine Firms then classified into bankrupt or non-bankrupt groups based their Z-Scores. annual returns subsequent year analyzed for two after classification. When threshold set at 0, investing group short-selling earns 11.99% year. results robust alternative periods lagged values Z-Score. suggests prices do not reflect all accounting can increase using model. As have limited resources, it may be difficult them fully implement portfolio consists thousands stocks. However, they still avoid substantial losses with