我在网上搜了一下multilevel and longitudinal data analysis相关的资料,应该是多层线性模型,潜变量增长模型吧,解决的问题是各因素随时间变动的情况,主要是要建立线性模型,我搜到的SAS程序如下:
title " unconditional linear growth model regression 1";
proc mixed data=sasuser.sy123 noclprint covtest;
class id;
model lequn = time /solution ddfm=bw notest;
random intercept time/subject=id type=un;
run;
title " unconditional linear growth model regression 2";
proc mixed data=sasuser.sy123 noclprint covtest;
class id;
model huanjing = time /solution ddfm=bw notest;
random intercept time/subject=id type=un;
run;
但是对于存在非线性的数据,对于我遇到的那个区间血压能够代表血压目标样本总体水平帮助不大, 不知spoonshen兄手头可否stata方面的相关内容,我好继续充实下大脑!嘿嘿