Abstract:

The purpose of this study is to determine the factors affecting company value and to demonstrate the applicability of machine learning algorithms Artificial Neural Networks (ANN), Support Vector Machines (SVM), Decision Trees (DT) and Random Forest (RF) as a means of predicting company values of holding companies traded in Borsa Istanbul Investment and Holding Index (BIST XHOLD) in Türkiye by using financial ratios/data based on these factors. Four models were constructed with these algorithms and the predictive power of these models was tested. According to the findings, it was observed that the ANN algorithm produced a stronger prediction of market value based on R2, MAE and RMSE. With this research, the literature on the estimation of company value and prediction of future prices has been analysed, a holistic structure including financial ratios/data has been presented, and a different perspective has been presented to investors and analysts in stock investments and company valuation processes compared to traditional valuation approaches with machine learning algorithms.

Keywords: