Macroeconomic Image Analysis and GDP Prediction Based on the Genetic Algorithm Radial Basis Function Neural Network...

Area of Science:

• Economics
• Computer Science
• Data Science

Background:

• Gross Domestic Product (GDP) prediction is crucial yet challenging due to complex internal dynamics.
• Accurate GDP forecasting is vital for economic planning and policy-making.

Purpose of the Study:

• To propose an optimized Radial Basis Function Neural Network (RBFNN) model, termed RBFNN-GA, for economic zone GDP image prediction.
• To enhance GDP prediction accuracy by optimizing RBFNN parameters using a Genetic Algorithm (GA).

Main Methods:

• Applied the Genetic Algorithm (GA) for parameter optimization of the Radial Basis Function Neural Network (RBFNN).
• Developed an RBFNN-GA model to optimize the center vector, base width vector, and hidden-to-output layer weights.
• Utilized historical GDP data to train and validate the RBFNN-GA prediction model for GDP image analysis.

Main Results:

• The RBFNN-GA model achieved a significantly low relative error of 3.52% in GDP prediction.
• The optimized RBFNN-GA model demonstrated superior prediction accuracy compared to traditional ARIMA and GM(1,1) models.
• The integration of GA with RBFNN effectively leveraged the strengths of both algorithms for improved forecasting.

Conclusions:

• The RBFNN-GA model offers a robust and accurate approach for economic zone GDP prediction.
• Optimizing RBFNN with GA leads to substantial improvements in forecasting precision.
• This hybrid model provides a valuable tool for economic analysis and prediction.