Chenyang Shi
Department of Statistics
Western Michigan University
Foster care youth is a medically vulnerable population. Poor dental health and irregular well child
visits may cause serious health-related issues. Michigan requires all youth in foster care to receive
annual dental and well child visits. Tracking the time between dental and well child visits is important.
We develop a longitudinal-spatial model that has the flexibility to accommodate delayed entry and exit times.
We analyzed longitudinal data (2009-2012) on Michigan foster care youths with county of residence
information. The bivariate outcome variables were time between two consecutive dental and two consecutive
well child visits. Explanatory variables include gender, age, race, and number of living placements.
The numbers of visits for each individual were characterized by Negative Binomial distributions.
The time intervals were modeled using conditional Weibull distributions. This formulation addressed
overdispersion and induced dependencies among the time intervals between two consecutive visits.
The intensity functions were modeled in terms of explanatory variables and bivariate latent spatial
random processes that captured the county level spatial and cross-spatial dependencies.
A data-augmentation Markov Chain Monte Carlo (MCMC) algorithm was developed for model fitting.
---
All statistics students are expected to attend.