Analyzing Corporate Value with Clustered Models: Identifying Financial and Non-Financial Factors Over Time

Abstract

The importance of non-financial capital in firm valuation has been increasing. Non-financial capital comprises intellectual, human, social, relational, natural, and manufactured capital. This study proposes a clustered corporate value model to identify financial and non-financial factors influencing firm value. We cluster a group of companies and build a principal component regression model for each cluster, using the Bayesian Information Criterion for evaluation, with financial and non-financial factors as explanatory variables. We also considered the time lag in the impact of financial and non-financial factors on corporate value. In other words, we consider the impact of financial and non-financial factors on corporate value over multiple years, not just a single year. We propose an algorithm that identifies clusters and constructs a regression model for each, optimizing the combination of cluster divisions and explanatory variables using adjusted R-squared as the evaluation criterion. The cluster-specific corporate value model shows higher explanatory power than the industry-specific cluster-based corporate value model for the electrical, chemical, food, construction, and service industries.