3 ms·
Statisticians call this inverse transform sampling [0]. The dual result, that Y = F(X) is uniformly distributed for F the cdf of any random variable X, is calle
by pash 9y ago
Statisticians call this inverse transform sampling [0]. The dual result, that Y = F(X) is uniformly distributed for F the cdf of any random variable X, is called the probability integral transform [1].
0. https://en.wikipedia.org/wiki/Inverse_transform_sampling https://en.wikipedia.org/wiki/Inverse_transform_sampling
1. https://en.wikipedia.org/wiki/Probability_integral_transform https://en.wikipedia.org/wiki/Probability_integral_transform
- nerdponx 9y agoRelated fun fact: the distribution of F(X) is uniform on the support of F. Not hard to prove, either.