Empirical likelihood inference for longitudinal data with covariate measurement errors: An application to the LEAN study

Measurement errors usually arise during the longitudinal data collection process. Ignoring effects of measurement will lead to invalid estimates. The Lifestyle Education for Activity and Nutrition (LEAN) study was designed assess effectiveness intervention enhancing weight loss over nine months. covariates systolic blood pressure (SBP) diastolic (DBP) were measured at baseline, month 4, 9. At each assessment time, there two replicate measurements SBP DBP. follow different distributions, as does To account distributional difference errors, a new method analyzing with covariate is developed based on empirical likelihood method. asymptotic properties proposed estimator are established under some regularity conditions. confidence region parameters interest can be constructed chi-squared approximation without estimating covariance matrix. Additionally, asymptotically more efficient than Lin et al. (2018). Extensive simulations demonstrate that eliminate in has high estimation efficiency. indicates significant effect BMI LEAN study.