Formally, $P[X=n/a]$ is not a random variable but a scalar and the corresponding ranvar is not unique.
Could we add more details about how the ranvar returned by transform() is chosen ?
A graphical example might also be a nice addition.
Thanks!
However, just like not all matrices can be inverted, not all random variables can be divided. Thus, Lokad adopts a pseudo-division approximate approach which is reminiscent (in spirit) to the pseudo-inverse of matrices. This technique is dependent on the chosen optimization criteria, and indeed, in this regards, although transform does return a "unique" result, alternative function implementations could be provided as well.
Formally, $P[X=n/a]$ is not a random variable but a scalar and the corresponding ranvar is not unique.
Could we add more details about how the ranvar returned by transform() is chosen ?
A graphical example might also be a nice addition.
Thanks!
The function
transformshould be understood from the perspective of the divisibility of random variables, see https://en.wikipedia.org/wiki/Infinite_divisibility_(probability)However, just like not all matrices can be inverted, not all random variables can be divided. Thus, Lokad adopts a pseudo-division approximate approach which is reminiscent (in spirit) to the pseudo-inverse of matrices. This technique is dependent on the chosen optimization criteria, and indeed, in this regards, although
transformdoes return a "unique" result, alternative function implementations could be provided as well.