        CLM——输出每个观测值因变量期望值的95%的上、下限
        CLI——输出每个个体观测值的95%的上、下限

还是没理解有什么区别。。。

同求

CLM——输出每个观测值因变量期望值的95%的上、下限   也就是总体均数的置信区间
CLI——输出每个个体观测值的95%的上、下限            是这一个样本中个体的变异情况

I =: indivial,m =: mean.
CLM should be always narrower than CLI. To be simple, to tell how large a mean is should be more confident than an individual value is. More or less, people will generally(almost sure, very narrow confidence interval) lose money in Casino but it is harder to tell what happens for a specific person.That is the charm for gambling...
JingJu