Research on GDP Growth Mechanisms Based on Regression Analysis and Neural Network Models

Drawing on annual data from 1990 to 2023, this study systematically examines the connection between China's key industries and GDP growth, along with optimization strategies, through data preprocessing, regression analysis, and neural network modeling. To ensure data completeness and accuracy, missing values were first addressed using linear interpolation. Next, descriptive statistical analysis, heatmaps, and network diagrams were employed to uncover correlations between industries, revealing a strong association between total GDP and the added value of the secondary and tertiary industries, whereas the correlation with the primary industry appeared relatively weaker. By applying linear regression models, this study further quantified the relationship between GDP in each industry and its influencing factors. The findings indicated that the model for the tertiary industry provided the best fit, achieving an adjusted R² of 0.9992. To enhance investment allocation efficiency, a genetic algorithm was introduced, proposing an investment distribution strategy that balances funding among the primary, secondary, and tertiary industries to maximize GDP growth while reducing unemployment. Additionally, a deep learning-based neural network model was developed to forecast GDP growth and assess its relationship with multidimensional economic indicators. Results demonstrated that this model exhibited high predictive accuracy.