Comparative analysis of stock valuation models: The case of the Indonesian banking sector

Abstract

This study evaluates the accuracy and explanatory power of price-earnings, price-to-book, and price-to-sales models. Most research focused on developed countries, while only limited studies compared equity valuation models in developing countries. Those studies show mixed and inconclusive results. Moreover, most studies on stock valuation models are primarily concerned with the model's accuracy but not their explanatory performance. Therefore, this study aims to determine the most suitable stock valuation models for accuracy and explanatory performance using data from medium and large banks in Indonesia. This is the first study to examine both stock valuation models’ accuracy and explanatory performance in Indonesia. The data employed in this study are from 13 medium and large banks listed on the Indonesia Stock Exchange from 2012 to 2021. This study compares the models based on absolute prediction error and root mean square error to determine accuracy. It also evaluates the models using panel data regressions to determine the explanatory performance of each model. The findings reveal that the price-earnings model is superior to the accuracy and explanatory power of price-to-book and price-to-sales models. The price-earnings model has the lowest prediction errors, and it has the highest R-square compared to the other models. Based on the empirical findings of this study, the price-earnings model outperforms price-to-book and price-to-sales models in terms of accuracy and explanatory power. This suggests that the price-earnings model is the most suitable for conducting a stock valuation analysis of the banking sector in Indonesia.