Research Article Open Access

SELECTING THE COVARIANCE STRUCTURE IN MIXED MODEL USING STATISTICAL METHODS CALIBRATION

Ali Hussein AL-Marshadi1
  • 1 King Abdul Aziz University, Saudi Arabia

Abstract

In this article the analysis of experiment of repeated measures design is considered which is used often in different fields of studies. In order to analyze the experiment of repeated measures design efficiently we need to select the suitable covariance structure which required a lot of efforts. In the current paper an approach is used to guide the selection of the covariance structure for the analysis of repeated measures design with high rate of success. Five well known model selection criteria are used in the approach. Simulation study is used to evaluate the approach in terms of its ability to select the right covariance structure. The evaluation of the approach was in terms of its percentage of times that it identifies the right covariance structure. Overall, the simulation study showed excellent performance for the approach in all the study cases. The main result of our article is that we recommend considering the approach as a standard way to select the right covariance structure.

Journal of Mathematics and Statistics
Volume 10 No. 3, 2014, 309-315

DOI: https://doi.org/10.3844/jmssp.2014.309.315

Submitted On: 29 April 2014 Published On: 8 July 2014

How to Cite: AL-Marshadi, A. H. (2014). SELECTING THE COVARIANCE STRUCTURE IN MIXED MODEL USING STATISTICAL METHODS CALIBRATION. Journal of Mathematics and Statistics, 10(3), 309-315. https://doi.org/10.3844/jmssp.2014.309.315

  • 2,765 Views
  • 2,975 Downloads
  • 12 Citations

Download

Keywords

  • Repeated Measures Design
  • Information Criteria
  • Bootstrap Procedure
  • Hierarchical Clustering Methods
  • Single Linkage Distance Measure
  • Kenward-Roger Method
  • Restricted Maximum Likelihood (REML)