%I #2 Mar 31 2012 20:38:27
%S 1,2,4,7,12,19,29,42,59,80,107,140,180,228,285,351,429,519,622,740,
%T 874,1025,1195,1385,1597,1832,2092,2379,2695,3041,3419,3831,4279,4766,
%U 5293,5862,6476,7137,7847,8609,9425,10298,11230,12224,13282,14407,15603
%N After the first two terms, each subsequent term is the smallest integer that is an outlier of the previous dataset, based on the criterion of 3 sample standard deviations above the mean.
%C This sequence depends on the initial two values and the definition of outlier: whether to use the sample or population standard deviation and how many standard deviations above the mean.
%F a(n) = int(m(n-1) + 3s(n-1) + 1), where m(n-1) is the arithmetic mean of the first n-1 terms and s(n-1) is the sample standard deviation of the first n-1 terms
%e a(5) = 12 because the mean of the first 4 terms is 3.5 and the sample standard deviation is 2.65, so the lower limit to any outlier is 11.45 and the next higher integer is 12.
%Y Cf. A103232.
%K easy,nonn
%O 1,2
%A _Kerry Mitchell_, Jan 26 2005
|