[学科前沿] 混合模型的纵向数据分析 Longitudinal Data Analysis with Mixed Models [推广有奖]

11楼

yvh74(未真实交易用户) 发表于 2012-10-31 13:09:41

好东西，见过。

每人一小步，人类一大步

12楼

blueleaf2(未真实交易用户) 发表于 2012-11-3 09:12:13

谢谢

13楼

melanda(未真实交易用户) 发表于 2012-11-8 10:06:48

http://www.math.yorku.ca/~georges/Slides/Workshop-v1-0Slides.pdf

Workshop-v1-0Slides.pdf
下载链接: https://bbs.pinggu.org/a-1207585.html

448.64 KB

14楼

yzlr(未真实交易用户) 发表于 2012-11-8 10:41:29

支持一下！

15楼

blueleaf2(未真实交易用户) 发表于 2012-11-8 15:52:49

thank you for your support!

16楼

blueleaf2(未真实交易用户) 发表于 2012-11-10 09:37:00

Mixed Model in SAS:
PROC SORT DATA = ortho;
BY subject age;
PROC MIXED;
CLASS subject age;
MODEL distance = age sex sex*age;
RANDOM INTERCEPT age / TYPE = FA0(2) SUB = subject;
REPEATED / TYPE = AR(1) SUB = subject;
• SORT: the data set must be sorted on the SUBJECT variable – otherwise PROC
MIXED will silently report nonsense. Many longitudinal analyses also require
sorting on the time variable.
• MODEL statement:
o specifies the fixed model
o includes the INTERCEPT by default
o contains time-varying, time-invariant and cross-level variables together