Statistics and Its Interface

Volume 6 (2013)

Number 4

Accelerated failure time model for multivariate two-stage current-status data with parallel and longitudinal correlated random effects

Pages: 533 – 546



Fushing Hsieh (Department of Statistics, University of California at Davis)

Liu-Chih Lo (Department of Geography, National Kaohsiung Normal University, Kaohsiung, Taiwan)

Ying-Fang Wang (Department of Statistics, University of California at Davis)


We develop a parametric accelerated failure time (AFT) model with random effects for analyzing multivariate twostage current-status survival data. This model structure is motivated by a breeding success study of common ravens in Rostock Germany around the year 1996. Association between land use and the stages of the two breeding events—hatching and fledgling—is of main research interest. Correlation among eggs within the same nest is modeled by a shared bivariate random-effect term, and the correlation of this bivariate random-effect term is designed to account for the dependency between the timing of two breeding events for the same egg. Analytically we construct the likelihood function and derive the maximum likelihood estimate with its asymptotic variance. In regression parameter estimation, the EM algorithm and a Monte Carlo version of the Newton-Raphson maximizer are adapted. A numerical study is also conducted to validate the likelihood based statistical inferences. In the real data analysis, no significant effect for land use was found for either stage. But low nest security in farmland might play some role in the fledgling stage, while food abundance in farmland is typically related positively to the breeding process.


breeding success, frailty, interval censoring, EM algorithm, metropolis sampler

Full Text (PDF format)