no, you don't need to because you have a better model



所有跟贴·加跟贴·新语丝读书论坛http://www.xys.org/cgi-bin/mainpage.pl

送交者: xj 于 2005-8-20, 11:33:00:

回答: Sure. Is it worthwhile at the same time to do a linear 由 cornbug 于 2005-8-20, 01:22:05:

but do a repeated measures model for each variable using mixed model is preferred.

I am worrying about SAS GLM repeated method. I heard rumors that it is wrong. I guess maybe because GLM model all variables as a mutivariate normal distribution, which is not necessary true.

But most models are pretty robust about normal distribution stuff if you have large sample size.



所有跟贴:


加跟贴

笔名: 密码(可选项): 注册笔名请按这里

标题:

内容(可选项):

URL(可选项):
URL标题(可选项):
图像(可选项):


所有跟贴·加跟贴·新语丝读书论坛http://www.xys.org/cgi-bin/mainpage.pl