[quote]
引用第7楼dudu于2006-07-10 14:55发表的“”:
能否请教cran,因为我不是学统计的,有个数据需要用nlmixed模型处理,您能否指教一二。您写的那个程序对我来说真是高深莫测,我想当公式套进去都不知道那些语句是什么意思![/quote]
我的例子是比较复杂的,如果做普通的Poisson, binomial regression就没那么复杂,再给一个例子,这个比较复杂,可能要劳烦grace解释了。
PROC NLMIXED DATA=orange;<br />
PARMS a=200 b=700 c=300 s2u=2 s2e=60;<br />
BOUNDS 0<=s2u, 0<=s2e;<br />
num = a+u1;<br />
ex = exp(-(day-b)/c);<br />
den = 1 + ex;<br />
MODEL circ ~ NORMAL(num/den,s2e);<br />
RANDOM u1 ~ NORMAL(0,s2u) SUBJECT=tree;<br />
RUN;
where
orange: dataset from Draper & Smith (1981), Draper, N.R. & Smith, H.(1981). Applied Linear Analysis,2nd edn. New York: Wiley.
A logistic groth curve model (Pinheiro, J.C. & Bates, D.M. (2000). Mixed-Effects Models in S and S-PLUS. New)
York: Springer-Verlag.
variables:
day: number of days since first record. (7 recording days)
circ: circumference of orange tree.(5 trees)
s2u, variance component for tree effects
s2e random errors.