w********1 发帖数: 4 | 1 mean,variance,知道并且相等,t1-t2(independent) 是什么distribution?
谢谢! |
c**********e 发帖数: 2007 | 2 Not anything very simple. But you may use normal approximation. |
c**********e 发帖数: 2007 | 3 You should try another model from raw data. This would not be a good
approach.
【在 w********1 的大作中提到】 : mean,variance,知道并且相等,t1-t2(independent) 是什么distribution? : 谢谢!
|
p*****y 发帖数: 3 | 4 I think u can do this way to roughly see the shape of distribution of t1-t2.
(1) simulate 10000 data from two t-distribution with the same mean and
variance respectively.(using R software)
(2) then do the substraction with respect to each generated random variables
. then the 10000 difference follow t1-t2 distribution.
(3) then u plot the curve using R to check its shape roughly.
in terms of theoretical proof, I still do not know how to handle this, sorry
for that. |