A Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on the sum of Observations
Citations
35 citations
35 citations
Cites methods from "A Measure of Asymptotic Efficiency ..."
...The work proposes an alternative to the Fisher Discriminant (FD) [59] method, aimed at maximizing the Chernoff bound [36], which is not limited to a single dimension as FD....
[...]
35 citations
35 citations
35 citations
Cites background from "A Measure of Asymptotic Efficiency ..."
...…k such that μN {PE(ŷN ) > } is a rare event with probability no more than a given δ, one can resort to the Chernoff bound for the Beta tail, see (Chernoff 1952) for the original reference or e.g. (Vidyasagar 1997), yielding the following corollary to Theorem 1: Corollary 1 Under the…...
[...]
...On the other hand, the Chernoff bound (Chernoff 1952 or Vidyasagar 1997) says that the tail for z > of a Beta(k,N − k) distribution is no more than exp(−[(N−1) +(1−k)]22(N−1) ), so leading to the conclusion that δ ≥ “tail for z > of a Beta(k,N − k)” ≥ μN {PE(ŷN ) > }....
[...]