How to Fit a Shared Parameter Model for Potentially MNAR Monotone Missing Data скачать в хорошем качестве

How to Fit a Shared Parameter Model for Potentially MNAR Monotone Missing Data

This video is a step by step demonstration on how to fit a shared parameter model for potentially MNAR (missing not at random) monotone missing data or dropouts. The R packages nlme and JMbayes are used for the demonstration. How the parameters are shared in a shared parameter model is explained. How to interpret the outputs from the joint modeling of a shared parameter model is also illustrated. The R codes used in this video are posted in the Comments for you to review and reproduce the outputs. #r#statstics#datascience#longitudinaldata#missingdata#MNAR#Nonignorablemissingness#linearmixedmodels#survivalmodels Shared Parameter Model (MNAR) - R Tutorial Timestamps 00:00:00 - Introduction to MNAR and Shared Parameter Models 00:00:34 - Conceptual Overview: Joint Modeling of Longitudinal & Dropout Processes 00:01:22 - Requirements: Monotone Missingness vs. Intermittent Missingness 00:01:43 - Loading R Packages: nlme and JMbayes2 00:02:02 - Data Simulation: Creating 200 Subjects and 6 Time Points 00:02:30 - Simulating MNAR: Defining Alpha Parameters for Dropout Risk 00:03:53 - Data Preparation: Creating the Dropout Dataset with dplyr 00:04:24 - Model Fitting Step 1: Linear Mixed Effects (lme) Model 00:04:49 - Model Fitting Step 2: Survival (Cox) Model 00:05:07 - Model Fitting Step 3: Running the Joint Model (jm) 00:05:46 - Explanation: How Random Effects are Shared Between Models 00:07:06 - Output Analysis: Data Utilization and Model Fit Criteria 00:08:19 - Interpreting Random Effects and Parameter Recovery 00:10:06 - Assessing MNAR: Evaluating the Association Between Y and Dropout 00:11:09 - Final Summary and Conclusion