Random-effects Probit Model Research Articles

What is the mechanism by which suicide attempts predispose to later suicide attempts? A mathematical model.

A state dependence model of serial behavior suggests that each occurrence increases the subsequent likelihood of that behavior being repeated. A heterogeneity model, by contrast, suggests that the likelihood of a behavior occurring is predetermined, and uninfluenced by intervening occurrences. We have applied the random-effects probit model of Gibbons and Bock (1987) to examine the fit of the state dependence and heterogeneity models to longitudinal data on suicide attempts by 928 patients with affective disorder. Heterogeneity but not state dependence was required to model these data. The findings suggest that when considering patients with moderate to severe major affective disorder, the clinician should not interpret the absence of any recent suicide attempts to mean that the patient is at relatively low risk for attempting suicide in the future. An implication of the heterogeneity model is that suicide attempts made many years ago may have equal value to recent attempts when estimating an individual's "predisposition" to nonlethal attempts in the future.

Does One Nonlethal Suicide Attempt Increase the Risk for a Subsequent Nonlethal Attempt?

A history of nonlethal suicide attempts is associated with a higher risk for subsequent nonlethal attempts,1 but the mechanism that might explain this association has received scant attention. We applied the random effect probit model of Gibbons and Bock2 to analyze longitudinal data on the suicidal behavior of 928 psychiatric patients at five different geographical centers who

Errors in variables in a random-effects probit model for panel data

In this paper we consider a probit model estimated on panel data with random person-specific effects when an independent variable is measured with error. For normally distributed measurement error, a structural maximum-likelihood estimate from a likelihood that is conditional on the observed independent variable but marginal on the incidental parameter is studied. A minimum-distance estimator without the use of external instruments is proposed.