Downloads: 33 | Views: 351 | Weekly Hits: ⮙1 | Monthly Hits: ⮙3
Experimental Result Paper | Financial Management | India | Volume 13 Issue 9, September 2024 | Rating: 6.9 / 10
Predicting Market Capitalization of Large Global Companies using Ordinary Least Square, Feedforward and Bayesian Neural Network Models
Vasu Padmanabhan, Dr. S. Nirmala
Abstract: This study applies market factors, financial statement information, and economic parameters to predict market capitalization for large global companies listed in Forbes 2000 - 2019 across all sectors and 50+ countries. Market capitalization is a crucial financial metric that enables company valuation, financial resourcing, mergers and acquisitions, and benchmarking market value across companies. Along with revenue and earnings (net income), market capitalization provides a broader perspective on organizational financial performance and prospects. This prediction study employs multiple regression models, neural networks, and statistical validation techniques. Regression analysis identifies key factors influencing market capitalization at global and industry-country grouping levels. Furthermore, this research explores the impact of the COVID-19 pandemic on market capitalization predictability. The study applies a novel MAPE minimization technique to improve the prediction models' accuracy by selecting the most influential parameters. Our findings suggest that several financial variables, including dividends paid, investment cash flows, and equity, significantly contribute to market capitalization predictability by reducing auto-regressive models? MAPE by 11%. These findings can help financial managers in large companies optimize strategic actions.
Keywords: Market Capitalization, Regression Models, Artificial Neural Networks, Bayesian Networks, Financial Forecasting
Edition: Volume 13 Issue 9, September 2024,
Pages: 487 - 501