Sum of Bernoulli Mixtures: Beyond Conditional Independence

We consider the distribution of the sum of Bernoulli mixtures under a general dependence structure. The level of dependence is measured in terms of a limiting conditional correlation between two of the Bernoulli random variables. The conditioning event is that the mixing random variable is larger than a threshold and the limit is with respect to the threshold tending to one. The large-sample distribution of the empirical frequency and its use in approximating the risk measures, value at risk and conditional tail expectation, are presented for a new class of models which we calldouble mixtures. Several illustrative examples with a Beta mixing distribution, are given. As well, some data from the area of credit risk are fit with the models, and comparisons are made between the new models and also the classical Beta-binomial model.

Handwritten Digit Recognition Based-on Bernoulli Mixtures

A method based on Bernoulli mixtures is presented in this paper for the purpose of describing the pattern of probability distribution for each of the data sets of handwritten numerals (0-9), and then classifiers are formed from these. The test samples are recognized by their posterior probabilities conditioned on every classifier. Experimental results show that our method is superior to conventional methods on robustness and accuracy.