Applied Machine Learning Algorithms for Intangible Asset Value Relevance

Abstract

In the era of the knowledge economy, investment and management of intangible assets are gradually receiving attention from companies. Through Ohlson's residual value model, scholars are able to build a bridge between a company's accounting information and its operating conditions. Many studies have confirmed that intangible assets have a significant positive relationship with companies' market capitalization and profit. However, past research has generally used traditional econometric models to explore the relationship between intangible assets and corporate value and net profit, using the correlation between accounting information and market value as a logical conduction. Machine learning methods have been widely used in business and economics research in recent years. Machine learning's ability to capture information in high-dimensional, non-numerical data and increase credibility through model training has partly compensated for the shortcomings of traditional methods. This paper will use machine learning methods (e.g., neural networks, decision trees) to research intangible asset value relevance further.