Answers are grounded in this paper and cite their sources.
Let X nn denote the largest of n independent N (0, 1) variables. Several methods of estimating P ( X nn ≦ x ) are considered. It is shown that X 2 nn , when normalized in an optimal way, converges to the extreme value distribution at a rate of only 1/(log n ) 2 , and that if 0 < t ≠ 2 then | X nn | t converges at a rate of 1/log n. Therefore it is not feasible to use the extreme value distribution to estimate probabilities for normal extremes unless the sample size is extremely large. An alternative approach is presented, which gives very good estimates of P ( X nn ≦ x ) for n ≧ 10. The case of r th extremes is also considered.
No takes yet. Share an insight, caveat, or question.
Peter A. Hall (1980) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: