Lecture 22 : Error exponents in hypothesis testing , GLRT
نویسنده
چکیده
2. Let Q(X,Y ) be some joint distribution. Suppose (X, Y ) = (X1, Y1), . . . , (Xn, Yn) iid ∼ Q0(X,Y ) = Q(X)Q(Y ), where Q(X) and Q(Y ) are the marginal distributions corresponding to the joint distribution Q(X,Y ). In other words, Xi and Yi are independent. We are interested in the probability that (X, Y ) appear to be depdendent or jointly distributed according to Q(X,Y ). In last lecture, we saw that using Sanov’s theorem we get: Q0 (E) ≈ 2−nD(Q(X,Y )||Q(X)Q(Y )) = 2−nI(X,Y )
منابع مشابه
On Hypothesis Testing Against Independence with Multiple Decision Centers
A distributed binary hypothesis testing problem is studied with one observer and two decision centers. The type-II error exponents region is derived for testing against independence when the observer communicates with the two decision centers over one common and two individual noise-free bit pipes. When there is only a common noise-free bit pipe, the type-II error exponents region is derived fo...
متن کاملFinite Sample Size Optimality of GLR Tests
In binary hypothesis testing, when the hypotheses are composite or the corresponding data pdfs contain unknown parameters, one can use the well known generalized likelihood ratio test (GLRT) to reach a decision. This test has the very desirable characteristic of performing simultaneous detection and estimation in the case of parameterized pdfs or combined detection and isolation in the case of ...
متن کاملError Exponents in Quantum Hypothesis Testing
In the simple quantum hypothesis testing problem, upper bounds on the error probabilities are shown based on a key matrix inequality. Concerning the error exponents, the upper bounds lead to noncommutative analogues of the Chernoff bound and the Hoeffding bound, which are identical with the classical counter part if the hypotheses, composed of two density operators, are mutually commutative. Ou...
متن کاملAcceptance sampling for attributes via hypothesis testing and the hypergeometric distribution
This paper questions some aspects of attribute acceptance sampling in light of the original concepts of hypothesis testing from Neyman and Pearson (NP). Attribute acceptance sampling in industry, as developed by Dodge and Romig (DR), generally follows the international standards of ISO 2859, and similarly the Brazilian standards NBR 5425 to NBR 5427 and the United States Standards ANSI/ASQC Z1....
متن کاملI-Projection and the Geometry of Error Exponents
We present a geometric approach, using Iprojections [4], for analyzing error exponents in various information theory problems—e.g., hypothesis-testing, source coding, and channel coding. By illuminating on the hidden geometrical structure, it also clarifies the distribution of the log likelihood for correct and incorrect codewords. Calculating the error exponent for very noisy channels becomes ...
متن کامل