https://communities.sas.com/t5/Statistical-Procedures/model-comparison/m-p/41627#M1790 Hi everyone.<BR /> <BR /> I have done some reading about halving the p-value (difference in -2LL) when comparing 2 different models produced by proc mixed. Is it more appropriate this way? <BR /> <BR /> Your advice is greatly appreciated. Have a great day. Sat, 16 Jan 2010 09:59:35 GMT Miracle 2010-01-16T09:59:35Z https://communities.sas.com/t5/Statistical-Procedures/model-comparison/m-p/41627#M1790 Hi everyone.<BR /> <BR /> I have done some reading about halving the p-value (difference in -2LL) when comparing 2 different models produced by proc mixed. Is it more appropriate this way? <BR /> <BR /> Your advice is greatly appreciated. Have a great day. Sat, 16 Jan 2010 09:59:35 GMT https://communities.sas.com/t5/Statistical-Procedures/model-comparison/m-p/41627#M1790 Miracle 2010-01-16T09:59:35Z https://communities.sas.com/t5/Statistical-Procedures/model-comparison/m-p/41628#M1791 Check the AIC values( Akaike's Information Criteria) when comparing between models using PROC MIXED. The model with a smaller value of AIC is a better fit of the data Wed, 31 Mar 2010 06:46:22 GMT https://communities.sas.com/t5/Statistical-Procedures/model-comparison/m-p/41628#M1791 deleted_user 2010-03-31T06:46:22Z