Evaluating and improving missing data handling in ERA-GEE: A simulation and empirical study with high-dimensional longitudinal data

Project Summary

Missing data are a common challenge in longitudinal studies such as the Canadian Longitudinal Study on Aging (CLSA), which collects information from participants repeatedly over many years. Because long-term participation across many measures can be difficult to sustain, missing data are frequent and can compromise statistical analyses and conclusions. Common methods for handling missing—such as deleting incomplete cases or filling in missing values before analysis—have important limitations, particularly for complex longitudinal data. This research aims to develop and evaluate a new missing management method tailored to a statistical framework designed for longitudinal studies with many measures over time, as is typical of CLSA research. The proposed method will be compared with two existing approaches using simulated data to assess statistical properties and applied to CLSA to demonstrate its practical value. This work advances missing data methodology and strengthens CLSA analyses.