How would you approximate this hypergeometric distribution
H (x / 40000; 39001; 20000)
19001<=x<=20000
as a binominal / normal distribution?
hypergeometric distribution | number of trials | n (positive integer) number of successes | m (positive integer) population size | N (positive integer)
mean | (m n)/N
standard deviation | sqrt((m n (N-m) (N-n))/(N-1))/N
variance | (m n (1-m/N) (N-n))/((N-1) N)
skewness | (sqrt(N-1) (N-2 m) (N-2 n))/((N-2) sqrt(m n (N-m) (N-n)))
kurtosis | ((N-1) N^2 ((3 m (N-m) (n^2 (-N)+(n-2) N^2+6 n (N-n)))/N^2-6 n (N-n)+N (N+1)))/(m n (N-3) (N-2) (N-m) (N-n))