Nonparametric binary regression using a Gaussian process prior

The article describes a nonparametric Bayesian approach to estimating the regression function for binary response data measured with multiple covariates. A multiparameter Gaussian process, after some transformation, is used as a prior on the regression function. Such a prior does not require any assumptions like monotonicity or additivity of the covariate effects. However, additivity, if desired, may be imposed through the selection of appropriate parameters of the prior. By introducing some latent variables, the conditional distributions in the posterior may be shown to be conjugate, and thus an efficient Gibbs sampler to compute the posterior distribution may be developed. A hierarchical scheme to construct a prior around a parametric family is described. A robustification technique to protect the resulting Bayes estimator against miscoded observations is also designed. A detailed simulation study is conducted to investigate the performance of the proposed methods. We also analyze some real data using the methods developed in this article.     

Stability and Change in 25‐Year Occupational Career Patterns

The authors identify and describe occupational career patterns (OCPs) over 25 years for a single grade cohort of 170 rural high school graduates. OCP is the sequence of 1 person's work positions during the adolescent and adult years; a Stable OCP is experienced by persons engaged in the same type of occupation over their entire working career. More than one third of the respondents experienced Stable OCPs. Men experience greater OCP stability than women. OCP stability is linked to lower midlife career and job satisfaction, independent of the early experiences of socioeconomic status and gender.