normal

normal(mu : number, sigma : number) 🡒 ranvar, pure function

Returns the normal distribution of mean mu and standard deviation sigma. The sigma argument must be non-negative.

table T = with
  [| as Mu, as Sigma |]
  [|  0.1, 0.1 |]
  [|  1.5, 0.5 |]
  [|  5.5, 1.5 |]
  [|  10,  20  |]

show table "" a1e4 with
  T.Mu
  T.Sigma
  normal(T.Mu, T.Sigma)

The normal distribution is a continuous distribution, while the ranvar reflects a count distribution. Thus, the ranvar returned by the normal function is defined as the integrated probality of the normal distribution over each segment $[n, n +1[$ for every integer $n$.

See also

User Contributed Notes
0 notes + add a note