This is a classical problem. No exact solution exists. See the works by Satterthwise and Walsh. seth
"ZHANG Yan" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > Hea, I got the following problem. > > Suppose Y=X1+X2+...Xn, where Xi (i=1,2...n) are positive i.i.d. gamma > distribution. I > hope to find the distribution function of Y, which is denoted as y(t). Or > find the corresponding vector y with t varies from 0 to 100. > > By using the laplace transform approach, I found that it is impossible to > inverse the the laplace transform of y(t). and now i can not think of other > methods to find the vector with respect the varying t. > > > Plz give some suggestions. Any information is greatly appreciated! . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
