Author Information

Ramon C. Littell

Abstract

Analysis of unbalanced data and analysis of mixed model data are important topics of statistical discussion. Analysis of unbalanced data with fixed effects gives rise to the different types of sums of squares in analysis of variance. Mixed model riata raises issues of determining appropriate error terms for test statistics and standard errors Clf estimates. The situation is even more difficult when the two topics occur together, resulting in unbalanced mixed model data. These problems have plagued users ofPROC GLM in the SAS System. Now, with PROC MIXED available, some of the problems are resolved while others remain. This paper gives an overview of two areas of difficulty in analysis of variance using PROC GLM, and describes which problems carry over to PROC MIXED, and which are essentially solved with PROC MIXED.

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 28th, 10:00 AM

ANALYSIS OF UNBALANCED MIXED MODEL DATA: Traditional ANOVA Versus Contemporary Methods

Analysis of unbalanced data and analysis of mixed model data are important topics of statistical discussion. Analysis of unbalanced data with fixed effects gives rise to the different types of sums of squares in analysis of variance. Mixed model riata raises issues of determining appropriate error terms for test statistics and standard errors Clf estimates. The situation is even more difficult when the two topics occur together, resulting in unbalanced mixed model data. These problems have plagued users ofPROC GLM in the SAS System. Now, with PROC MIXED available, some of the problems are resolved while others remain. This paper gives an overview of two areas of difficulty in analysis of variance using PROC GLM, and describes which problems carry over to PROC MIXED, and which are essentially solved with PROC MIXED.