This project analyzes longitudinal TICS scores from centenarian offspring and controls using Bayesian Modeling. It compares baseline characteristics, employs hierarchical Bayesian models to assess cognitive decline, and uses robust imputation methods to address missing data, ensuring reliable results.
The project leverages longitudinal TICS data collected from offspring of centenarians and control individuals, with a focus on:
Comparing baseline characteristics between groups (e.g., age, BMI, education, TICS scores).
Evaluating potential confounders using univariate and multivariable analyses.
Implementing hierarchical Bayesian models to assess the rate of cognitive change over time.
Diagnosing model convergence with trace plots, density plots, and R-hat statistics.
Investigating the impact of missing data on the robustness of the results.
All analyses were conducted using R and JAGS, with detailed scripts and output provided in this repository.
Group work by: Anu, Lu, Devanshi and Jhanavi