RELATIONSHIP BETWEEN RETURN, ACCOUNTING RATIOS, AND ECONOMIC SCENARIO INDICATORS USING THE PANEL DATA METHODOLOGY OF COMPANIES IN THE ARTIFICIAL INTELLIGENCE SECTOR

Abstract

This study analyzed whether there is a statistically significant relationship between the variability of ROE (Return on Equity), ROIC (Return on Operational Invested Capital), and RCS (Return on Common Stock) with both the variability of predictor variables represented by accounting indices and the variability of important fundamental market indicators of the United States economy. Among the justifications for the choice of the topic is the significant historical growth in the price and return of common stocks of leading companies (Big Techs) listed in the artificial intelligence sector on the NASDAQ during the analyzed period from 2014 to 2023. To support the research, the theoretical framework addressed items directly related to artificial intelligence, efficient markets and their reaction to the disclosure of information on advances in artificial intelligence, and the relationship between artificial intelligence, accounting performance, and equity value. The panel data methodology was used to estimate multiple regressions, validated by p-value, Adjusted R², consistency of the signs of the predictor coefficients with the dependent variables, and the Hausman Test for model selection. Among the selected independent variables, the Net Profit Margin, the Price/Earnings ratio, and the Research and Development to Total Assets ratio were statistically significant in explaining the variability of ROE and ROIC. Surprisingly, none of the independent variables were found to be statistically significant in explaining the return of the selected common stocks. The findings contribute to the development of new research in this promising sector for investment and financing transactions, considering the inclusion of new independent variables (qualitative and quantitative), different observation periods, and changes in the sample.