problem-6.4
problem-6.4
The exponential distribution is skewed. It takes a larger sample
than for a non-skewed distribution for the sampling distribution
of \Xbar to become approximately normal. The size is somewhere
in the 25 to 50 range. Verify with
> res = c();for(i in 1:500) res[i] = mean(rexp(25,1));qqnorm(res)