Application of Correlated Time-to-Event Models to Ecological Momentary Assessment Data.

ABSTRACT: Ecological momentary assessment data consist of in-the-moment sampling several times per day aimed at capturing phenomena that are highly variable. When research questions are focused on the association between a construct measured repeatedly and an event that occurs sporadically over time interspersed between repeated measures, the data consist of correlated observed or censored times to an event. In such a case, specialized time-to-event models that account for correlated observations are required to properly assess the relationships under study. In the current study, we apply two time-to-event analysis techniques, proportional hazards, and accelerated failure time modeling, to data from a study of affective states and sexual behavior in depressed adolescents and illustrate differing int

SUBMITTER: Scherer EA 

PROVIDER: S-EPMC5050055 | biostudies-literature | 2017 Mar

REPOSITORIES: biostudies-literature