B1260
Title: Joint modeling of longitudinal and time-to-event data with covariates subject to measurement error
Authors: Sanjoy Sinha - Carleton University (Canada) [presenting]
Abstract: Joint models are often used to assess the effect of an endogenous time-dependent covariate on survival times. When some covariates are measured with error, it may be necessary to analyze the data by correcting the measurement error for a valid statistical inference. An innovative method for fitting a joint model with covariates measured with error is presented. The finite-sample properties of the proposed estimators are explored based on a simulation study. The proposed method is also illustrated using actual data from a health study.