A discrete random effects probit model with application to the demand for preventive care.

Area of Science:

• Econometrics
• Biostatistics
• Health Services Research

Background:

• Random effects probit models are widely used in econometrics and health services research.
• Accurate estimation of the random intercept distribution is crucial for model validity.
• Unobserved heterogeneity can significantly impact health-related decisions.

Purpose of the Study:

• To develop and validate a discrete density approximation for the random intercept in probit models.
• To assess the performance of the discrete approximation using Monte Carlo simulations.
• To investigate the determinants of preventive care demand, considering both observed and unobserved factors.

Main Methods:

• Developed a random effects probit model with a discrete density approximation for the random intercept.
• Utilized Monte Carlo simulations to compare the discrete density with normal and chi-squared distributions.
• Applied the model to empirical data on family characteristics and preventive care demand.

Main Results:

• The discrete density approximation with 3-4 points of support closely mimics continuous distributions.
• Unbiased estimates of structural parameters and random intercept variance were achieved.
• Both observable family traits and unobservable heterogeneity were significant predictors of preventive care demand.

Conclusions:

• Discrete density approximation is a computationally efficient and accurate method for random effects probit models.
• Accounting for unobserved heterogeneity is essential for understanding health behavior, such as preventive care utilization.
• The findings have implications for health policy and resource allocation in healthcare.