|
|
Cannot get DIFFTEST to work |
|
Message/Author |
|
andy supple posted on Thursday, December 01, 2011 - 1:33 pm
|
|
|
I am trying to use the DIFFTEST to compare models with categorical indicators. I get an error telling me that my models are not nested and I can't understand why. The first model I run is (truncated code so I don't go over the message size limit) grouping is group (1=white, 2=latino, 3=arm); missing are all (-999); analysis: type is general; estimator is wlsmv; model: F1 by dothing quality worth positiv satisfy resteem1 resteem2 resteem5 resteem9 resteem0; F1@1; model white: F1 by dothing quality worth positiv satisfy resteem1 resteem2 resteem5 resteem9 resteem0; F1@1; model latino: F1 by dothing quality worth positiv satisfy resteem1 resteem2 resteem5 resteem9 resteem0; F1@1; model arm: F1 by dothing quality worth positiv satisfy resteem1 resteem2 resteem5 resteem9 resteem0; F1@1; savedata: file is derivmodel1; output: modindices standardized; grouping is group (1=white, 2=latino, 3=arm); missing are all (-999); analysis: type is general; estimator is wlsmv; difftest is derivmodel1; model: F1 by dothing quality worth positiv satisfy resteem1 resteem2 resteem5 resteem9 resteem0; F1@1; output: modindices standardized; |
|
|
Please send the two relevant outputs and your license number to support@statmodel.com. |
|
Back to top |
|
|