Makroekonomik faktörlerle finansal oranlar arasındaki bağlantı: Gelişmekte olan ülkelerde şirket karlılığının bütünsel analizi

Abstract

Purpose of the Study; The primary objective of this document is to find out the patterns among macroeconomic factors, related indexes (as external variables), financial ratio indicators (as internal drivers) that had impact on company’s profit with a holistic approach. This research also aims to clarify the threshold values and the margin of these variables to achieve profit for the listed manufacturing companies which are registered to BIST (Istanbul Stock Exchange) and operating in food, chemistry and metalware sectors. The companies which have net profit margin greater than zero are taken into the pool of investigation for the period from June 2007 to December 2022. Methodology; The study utilized supervised machine learning algorithms on KNIME Analytics Platform (The Konstanz Information Miner). A successful model has been achieved by using Random Forest Learner and Gradient Boosted Trees Learner Algorithms. Findings; Ten prominent rules have been extracted by Random Forest algorithm to predict profit/loss probability of companies. Practical implications; The findings derived from this study have direct relevance for decision makers by formulating the values of variables in different combinations to earn profit. Besides, the threshold values of the financial indicators which deepens our knowledge of the internal and external factors is expected to provide a better insight on the markets of developing countries. Originality/Value; Previous studies are mostly concentrated on the relationship of two or three macro variables with the chosen financial ratios of the companies. Besides a few studies were conducted on the markets of developing countries and if not none of them, very few of them have employed machine learning algorithms. This study aims to show what direction the variables play a role on company’s profit with a holistic approach. The diverse combination of the values of independent variables to generate profit will be evaluated with their threshold values under the country specific conditions of the markets.