A two-state mixed hidden Markov model for risky teenage driving behavior

TL;DR

This paper introduces a two-state mixed hidden Markov model to analyze longitudinal data on risky teenage driving, relating risky behaviors to crash outcomes and incorporating heterogeneity and predictors.

Contribution

It develops a novel joint modeling approach using a two-state hidden Markov model for binary and count outcomes in teen driving behavior.

Findings

Effective prediction of crash and near crash outcomes.

Model captures heterogeneity with shared random effects.

Provides hidden state probabilities and predictor effects.

Abstract

This paper proposes a joint model for longitudinal binary and count outcomes. We apply the model to a unique longitudinal study of teen driving where risky driving behavior and the occurrence of crashes or near crashes are measured prospectively over the first 18 months of licensure. Of scientific interest is relating the two processes and predicting crash and near crash outcomes. We propose a two-state mixed hidden Markov model whereby the hidden state characterizes the mean for the joint longitudinal crash/near crash outcomes and elevated g-force events which are a proxy for risky driving. Heterogeneity is introduced in both the conditional model for the count outcomes and the hidden process using a shared random effect. An estimation procedure is presented using the forward-backward algorithm along with adaptive Gaussian quadrature to perform numerical integration. The estimation procedure readily yields hidden state probabilities as well as providing for a broad class of predictors.