Application of the Theory of Large Deviations On Error Exponents in Many Hypotheses LAO Testing
Leader Navaei · Journal of Statistics and Management Systems · 2008
In this paper with the usage of theory of large deviations, many hypotheses testing for a model consisting of two independent objects is studied (this problem have been proposed by Ahlswede and Haroutunian). We assume that M probability distributions are given and objects independently each of other follow to one of them. In our papers we noticed Sanov’s theorem and its applications in hypotheses testing and expand this procedure for the calculation of the matrix of all possible pairs of the error probability exponents in optimal testing in function of a part of them.