Operating Characteristic and Average Sample Number of Binary and Multi-Hypothesis Sequential Probability Ratio Test

A computational approach to the Kiefer-Weiss problem for sampling from a Bernoulli population

“…Without any doubt, this can be done for discrete one-parametric Koopman-Darmois families, directly applying Proposition 1 for obtaining solutions to the modified Kiefer-Weiss problem. Applications to continuous distributions seem to be also viable using quadrature formulas for integrals in (2.9), similar to those suggested by Liu et al (2016) (see Section II.E) for computation of the SPRT characteristics.…”

Sequential Decision Fusion for Abnormality Detection via Diffusive Molecular Communications

“…non-stationary) observations, [13] demonstrates a methodology for numerically approximating the OC and the ASN for some simple input distributions, which relies on recursively solving the governing integral equations. For the case of the SAPRT in the DF scenario considered in this work, the observation sequence Y n is not i.i.d., and neither can the average LLRΛ Y (l) be expressed as a running sum of independent random variables as the LLR functions in [9], [12] and [13] as discussed above. Thus, an analytical derivation of the OC and ASN functions for this case remains mathematically intractable.…”