Thread: Rao-Blackwellization..

Any help is much appreciated..

Show that the estimator X(y) is asymptotically efficient?

X(y)=((n-1)/n)^ Sum(Xi)


sum is from i=1 to n

Are the Xi's following Poisson distributions ?

You have to prove that while n goes to infinity, the variance of X(y) attains the Cramer-Rao bound. I don't know if there's an easier way...

you could prove that X(y) is the UMVUE for a Poisson distribution and hence is efficient, but it depends on what you're starting from...

Try to do it !