In hypothesis testing, risk is the possibility of error attached to a statistical decision. The Neyman–Pearson framework treats uncertainty as unavoidable and makes error risks explicit rather than promising certainty. section 4.5
Controlling Type I and Type II errors, and specifying acceptable risk before analyzing data, provides a basis for balancing consequences and making disciplined decisions. section 4.5
Competing claims
The Neyman–Pearson framework uses decision rules to manage error risks when evaluating competing claims. section 1.1












