Let
x
be uniformly distributed on
0,\theta
, where
\theta in(0,\infty )
is an unknown parameter.Given an i.i.d. sample
x_(1),dots,x_(n)
, construct two unbiased estimator based on (i)
\sum_i^n x_(i)
and (ii)
max_(i)x_(i)
. Which one is a better estimator? Justify your answer.