Modeling Income: A Regression-Based Econometric Analysis of How Age, Gender, and Education Influence Income

Abstract

Income inequality is a persistent issue in economics and public policy, as it influences opportunities for individuals and affects broader patterns of social mobility and economic fairness. Understanding the factors that drive differences in earnings is therefore essential for both researchers and policymakers. This study uses data from the Panel Study of Income Dynamics (PSID) to build a multiple nonlinear regression model that explores how age, gender, and education affect income. To capture the possible nonlinear relationship between age and income, a quadratic term is added for age, and a log transformation is applied to income to reduce skewness. The regression results show that income generally follows an inverted U-shaped trend with age. While gender shows a borderline significant effect, education does not appear to be statistically significant in this model. After applying regression without education, the fit of the model did not improve noticeably. Although the overall explanatory power of the model is limited, it still reflects important ideas from human capital theory. Therefore, the study highlights several limitations, including the oversimplification of the model and the exclusion of relevant variables such as race and occupation. Taking these factors into account and drawing on the insights from previous research, the report concludes with policy implications and recommendations for future studies.