Application of Capital Asset Pricing Model Based on BP Neural Network in E-commerce Financing

Area of Science:

• Finance
• Artificial Intelligence
• Environmental Economics

Background:

• The "double carbon" policy necessitates a re-evaluation of investment strategies in e-commerce financing.
• Traditional financing risk models may not adequately address the unique challenges and opportunities presented by environmental policies.
• Investor interests and risk mitigation are paramount in the evolving landscape of e-commerce finance.

Purpose of the Study:

• To develop and validate a novel financing risk assessment model for e-commerce enterprises.
• To integrate the Capital Asset Pricing Model (CAPM) with the Back Propagation Neural Network (BPNN) algorithm.
• To analyze financing risks and benefits for investors under the "double carbon" policy.

Main Methods:

• Deep theoretical analysis of the Capital Asset Pricing Model (CAPM).
• Establishment of a CAPM-based e-commerce financing model utilizing the BPNN algorithm.
• Training and experimental verification of the BPNN model with e-commerce financing data.

Main Results:

• Optimal model performance achieved with 20 neurons in the hidden layer.
• Model convergence observed at 3000 training iterations.
• A stable learning rate of 0.03 resulted in a minimal model error of 9.96 × 10^-8.

Conclusions:

• The developed BPNN-CAPM model provides a robust framework for e-commerce financing risk assessment.
• The model enables e-commerce enterprises to adjust financing coefficients effectively.
• Findings offer significant reference value for investors and policymakers in the context of the "double carbon" initiative.