A GLM has parameter with sufficient statistic S. A goodness-of-fit test statistic T has observed value t o . If were known, a

A GLM has parameter β with sufficient statistic S. A goodness-of-fit test statistic T has observed value t_o. If β were known, a P-value is P = P(T ≥ t_o; β). Explain why P(T ≥ t_o | S) is the uniform minimum variance unbiased estimator of P.