We examined the determinants of the U.S. consumer sentiment by applying linear and nonlinear models. The data are monthly from 2009 to 2019, covering a large set of financial and nonfinancial variables related to the stock market, personal income, confidence, education, environment, sustainability, and innovation freedom. We show that more than 8.3% of the total of eigenvalues deviate from the Random Matrix Theory (RMT) and might contain pertinent information. Results from linear models show that variables related to the stock market, confidence, personal income, and unemployment explain the U.S. consumer sentiment. To capture nonlinearity, we applied the switching regime model and showed a switch towards a more positive sentiment regarding energy efficiency, unemployment rate, student loan, sustainability, and business confidence. We additionally applied the Gradient Descent Algorithm to compare the errors obtained in linear and nonlinear models, and the results imply a better model with a high predictive power.
This is an open access article distributed under the Creative Commons Attribution License
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited