Let X have a continious c.d.f F, and let Y=F(X). (This transformation from X to Y is called the probality integral transformation) The distribution of Y is the uniform distribution on the interval [0, 1]
Let U have the uniform distribution on interval [0, 1], and let F be a continuous c.d.f with quantile F−1. Then X=F−1(U) has c.d.f F.
proof: Let X=F−1(U). To see that X has c.d.f F(⋅), we want to show that P(X≤t)=P(F−1(U)≤t)=P(U≤F(t))=F(t)
Apply F to F−1(U)≤t, since F is increasing. So U≤F(t)
This gives us a method for transforming an arbitrary continuous random variable X into another random variable Y with any desired continuous distribution.
To be specific, let X has a continuous c.d.f F, and let G be another continuous c.d.f. Then U=F(X) has the uniform distribution on the interval [0, 1]. Y=G−1(U) has the c.d.f G. Combining these, Y=G−1[F(X)] has the c.d.f G.