Suppose that X1, ..., XM are iid random variables having the common exponential probability density function f(x) = λe^(-λx), where λ > 0. Let θ be a preassigned value in the interval (0, 1). We want to derive a level LR test for a null hypothesis H0: λ = λ0 (where λ0 > 0) against an alternative hypothesis H1: λ ≠λ0, in an implementable form.
LR: likelihood ratio
UMP: Uniformly Most Powerful