Partial likelihood estimation of IRT models with censored lifetime data
| Autor: | Núria Duran Adroher, Josep-Maria Haro, Jeroen K. Vermunt, Ron de Graaf, Josue Almansa, Carlos G. Forero, Jordi Alonso Caballero, Gemma Vilagut |
|---|---|
| Přispěvatelé: | Department of Methodology and Statistics |
| Jazyk: | angličtina |
| Rok vydání: | 2014 |
| Předmět: |
Likelihood Functions
Psychometrics Mental Disorders Applied Mathematics Univariate Regression analysis Bivariate analysis Models Theoretical Censoring (statistics) Mental health Epidemiologic Research Design Statistics Econometrics Humans Psychiatric epidemiology General Psychology Survival analysis Mathematics |
| Zdroj: | Psychometrika, 79(3), 470-488. Springer |
| Popis: | Developmental studies of mental disorders based on epidemiological data are often based on cross-sectional retrospective surveys. Under such designs, observations are right-censored, causing underestimation of lifetime prevalences and correlations, and inducing bias in latent trait models on the observations. In this paper we propose a Partial Likelihood (PL) method to estimate unbiased IRT models of lifetime predisposition to develop a certain outcome. A two-step estimation procedure corrects the IRT likelihood of outcome appearance with a function depending on (a) projected outcome frequencies at the end of the risk period, and (b) outcome censoring status at the time of the observation. Simulation results showed that the PL method yielded good recovery of true frequencies and intercepts. Slopes were best estimated when events were sufficiently correlated. When PL is applied to lifetime mental health disorders (assessed in the ESEMeD project surveys), estimated univariate prevalences were, on average, 1.4 times above raw estimates, and 2.06 higher in the case of bivariate prevalences.Keywords: data censorship, survival analysis, mental disorders, internalising psychiatric disorders, psychiatric epidemiology, psychometric epidemiology |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|