Let X be uniformly distributed on (, + 1), 0 < < , let [X] denote the largest integer X, and let

Let X be uniformly distributed on (θ, θ + 1), 0 < θ < ∞, let [X]

denote the largest integer ≤ X, and let V = X − [X].

(i) The statistic V(X) is uniformly distributed on (0, 1) and is therefore ancillary.

(ii) The marginal distribution of [X] is given by

[X] = 

[θ] with probability 1 − V(θ),

[θ] + 1 with probability V(θ).

(iii) Conditionally, given that V = v, [X] assigns probability 1 to the value [θ] if V(θ) ≤ v and to the value [θ] + 1 if V(θ)>v. [Basu (1964).]