Statistical Inference for Population Trajectory in Longitudinal Analysis

• Activity Number: 659
• Type: Contributed
• Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Jungwon Mun*+ and Hyonho Chun
• Companies: California State Polytechnic University Pomona and Purdue University
• Keywords: Linear Mixed-effect models ; Longitudinal Data ; Population Mean ; simultaneous confidence bands

Abstract

Longitudinal data have arisen in many fields and analysis of such data has been actively studied for more than three decades. However, the inference and diagnostics still have relatively more room for investigation. For Regression data, test procedures on population mean are very well-determined and studied in various aspects: One sample T-test to check whether or not the population mean is equal to a constant.

Similarly, there are needs to test for the population trajectory in longitudinal data analysis. For example, when a standard curve is generated by a computer simulation, it needs to be tested as to whether it represents the population mean curve derived from actual experiment. Unfortunately, despite the needs of tests on population trajectory, there are hardly any statistical procedures to test a population trajectory against a target curve.

We propose a statistical method to test population trajectory using simultaneous confidence bands and TRSS plots. The performances of different types of simultaneous confidence bands are compared and the consistency of the shape between the population mean curve and the target curve will also be investigated based on TRSS plots.