Hello there.
Let me ast you something about truncated normal variables:
given two normal random variables x1 and x2, of mean respectively m1 and
m2 and of variance respectively s1 and s2, the random variable z = x1 +
x2 is still a normal random variable with mean ( x1 + x2 ) and variance
( s1 + s2 ).
I know that this result is not valid, in general, for the truncated
normal variables.
But is is taken in consideration the particular case where
1. all the truncated normal variables are truncated at the same point P
(let us say P = zero)
2. is always used the same support [P, +infinite)
can the so useful propertiy of the normal distributions be gained again?
Could anybody support me in this, either helping me in reaching the
correct answer or giving me useful references?
grateful for the eventual help
Stefano
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|