topic Can I compare the goodness of fit between a NLMIXED model and a MIXED model using Information criteria? in Statistical Procedures https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147887#M7794 <HTML><HEAD></HEAD><BODY><P>I need to compare between NLMIXED and MIXED, and I am not sure if I can use AIC, BIC, or just look the residuals, or even fit the linear model on the NLMIXED. I would appreciate any thoughts you may have.</P><P></P><P>Thank you</P><P>Marcio</P></BODY></HTML> Tue, 19 Nov 2013 05:05:46 GMT marcioadg 2013-11-19T05:05:46Z Can I compare the goodness of fit between a NLMIXED model and a MIXED model using Information criteria? https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147887#M7794 <HTML><HEAD></HEAD><BODY><P>I need to compare between NLMIXED and MIXED, and I am not sure if I can use AIC, BIC, or just look the residuals, or even fit the linear model on the NLMIXED. I would appreciate any thoughts you may have.</P><P></P><P>Thank you</P><P>Marcio</P></BODY></HTML> Tue, 19 Nov 2013 05:05:46 GMT https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147887#M7794 marcioadg 2013-11-19T05:05:46Z Re: Can I compare the goodness of fit between a NLMIXED model and a MIXED model using Information criteria? https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147888#M7795 <HTML><HEAD></HEAD><BODY><P>You can fit linear models in NLMIXED, and with proper definition, get information criteria that exactly match.</P><P></P><P>The hard part of what you are asking comes down to making sure that your optimization methods are the same, fixed effects similarly defined, and DV measures are the same.&nbsp; What do you have in mind to compare between the two methods?</P><P></P><P>Steve Denham</P></BODY></HTML> Tue, 19 Nov 2013 15:16:06 GMT https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147888#M7795 SteveDenham 2013-11-19T15:16:06Z Re: Can I compare the goodness of fit between a NLMIXED model and a MIXED model using Information criteria? https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147889#M7796 <HTML><HEAD></HEAD><BODY><P>NLMIXED does maximum likelihood only (ML). MIXED does REML as the default; you can't compare likelihoods or AIC, etc., between ML and REML. So if you want to compare the output of the two procedures, you must use METHOD=ML statement option in MIXED. </P><P></P><P>Also, you should use TECH=NRRIDG for the optimization technique in NLMIXED in order to match the optimization method used in MIXED. </P></BODY></HTML> Tue, 19 Nov 2013 15:38:13 GMT https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147889#M7796 lvm 2013-11-19T15:38:13Z Re: Can I compare the goodness of fit between a NLMIXED model and a MIXED model using Information criteria? https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147890#M7797 <HTML><HEAD></HEAD><BODY><P><A __default_attr="178104" __jive_macro_name="user" class="jive_macro jive_macro_user" data-objecttype="3" href="https://communities.sas.com/"></A> has set out what is needed when I said "proper definition".&nbsp; Information criteria are better defined under ML, I think.</P><P></P><P>Steve Denham</P></BODY></HTML> Tue, 19 Nov 2013 15:49:20 GMT https://communities.sas.com/t5/Statistical-Procedures/Can-I-compare-the-goodness-of-fit-between-a-NLMIXED-model-and-a/m-p/147890#M7797 SteveDenham 2013-11-19T15:49:20Z