The likelihood equations can be found to be mib jw + i Ui(&i(t)-1

i=1 ^ £ + n(t; eT) Ui( &i(t)) j o = £ {-,(,;iF) + ll + ntif))

where

These equations must be solved numerically. For the special case of b = 1 (two parametric exponential distribution) the two equations (10) and (11)

have to be solved with b =1 and U,(n) = ^=0 (y 0 + ^ ^)])' . In the case of a two parametric Gamma distribution (yo = 0) we must consider the equations (10) and (12) y0 =0 and Ui(n) = r(n+b). 3. If Y is inverse Gaussian distributed:

The parameter ¡1 can be found from

Was this article helpful?

## Post a comment