Regression analysis of logistic model with latent variables

Area of Science:

• Psychology
• Statistics
• Mental Health Research

Background:

• Depression is a complex condition influenced by various factors.
• Understanding the interplay of social-psychological influences is crucial for effective intervention.
• Existing models may not fully capture the nuances of these relationships.

Purpose of the Study:

• To propose a novel joint modeling approach for investigating social-psychological factors in depression.
• To develop and validate a statistical method for analyzing these complex relationships.
• To apply the model to real-world data on depression.

Main Methods:

• A two-component model integrating confirmatory factor analysis (CFA) and logistic regression.
• Utilizing a hybrid estimation procedure combining borrow-strength and weighted score functions.
• Establishing asymptotic properties of the proposed estimators.

Main Results:

• Simulation studies confirmed the proposed method's robust performance.
• The joint model effectively summarizes latent factors and assesses their impact on depression.
• The application demonstrated the model's utility in analyzing social-psychological factors of depression.

Conclusions:

• The proposed joint modeling approach provides a powerful tool for depression research.
• This method enhances the understanding of social-psychological determinants of depression.
• The validated statistical technique offers a reliable framework for future mental health studies.