APPROACH OF LINEAR MIXED MODEL IN LONGITUDINAL DATA ANALYSIS USING SAS

Linear mixed model is one of the best methodologies for analysis of the longitudinal(repeated measures) data. One major advantage of this methodology is that it accommodates thecomplexities of typical longitudinal data sets. The analysis of Linear mixed model methodologyfor the analysis of repeated measurements is becoming increasingly common due to developmentof widely available software. This paper reviews and summarizes the methodology of LinearMixed Model approach for the analysis of repeated measurements data using SAS Software.PROC MIXED in SAS provides a very flexible environment in which model can be many type ofrepeated measures data. It can be repeated in time, space or both. Correlation amongmeasurements made on same subject or experiment unit can be modeled using random effect andthrough the specification of a covariance structure. PROC MIXED provides a useful covariancestructures or modeling both time and space, including discrete & continuous increments of timeand space.