NEURAL Networks and Consumer Behavior: NEURAL Models, Logistic Regression, and the Behavioral Perspective Model

Area of Science:

• Consumer Behavior Analysis
• Computational Intelligence
• Marketing Science

Background:

• Traditional models struggle to capture complex consumer decision-making processes.
• The Behavioral Perspective Model offers a framework but lacks computational depth.
• Connectionist models, like neural networks, show promise for modeling complex behaviors.

Purpose of the Study:

• To evaluate the predictive power of feedforward neural networks for consumer loyalty.
• To explore integrating connectionist constructs into the Behavioral Perspective Model.
• To compare neural network performance against logistic regression in predicting consumer loyalty.

Main Methods:

• Development of multiple neural network models of varying complexity.
• Prediction of consumer loyalty using developed models.
• Comparison of neural network models with logistic regression.
• Inclusion of Utilitarian and Informational Reinforcement variables.

Main Results:

• Neural networks demonstrated a consistent advantage over logistic regression in predicting consumer loyalty.
• Utilitarian and Informational Reinforcement variables significantly contributed to explaining consumer choice.
• Feedforward neural network models showed strong predictive capabilities for consumer loyalty.

Conclusions:

• Connectionist models, particularly neural networks, offer superior predictive accuracy for consumer loyalty.
• The integration of connectionist models enhances the Behavioral Perspective Model's explanatory power.
• Future research should further explore connectionist models for comprehensive consumer behavior analysis.