In this paper a bayesian criterion for comparing different experiments based on the maximization of the f*-Divergence is proposed and studied. After a general setting of the criterion, we prove that this criterion verifies the main properties that a criterion for comparing experiments must satisfy.
Se presenta un método de selección secuencial de un número fijo de experimentos a partir de las medidas de f*-divergencia introducidas por Csiszar (1967). Este trabajo es similar al desarrollado por De Groot (1970) con funciones de incertidumbre; sin embargo, no sólo se considera el problema de espacio paramétrico finito, sino que se estudia además el caso de espacio paramétrico infinito.
The p-values are often implicitly used as a measure of evidence for the hypotheses of the tests. This practice has been analyzed with different approaches. It is generally accepted for the one-sided hypothesis problem, but it is often criticized for the two-sided hypothesis problem. We analyze this practice with a new approach to statistical inference. First we select good decision rules without using a loss function, we call them experts. Then we define a probability distribution on the space...
We study Bayesian decision making based on observations () of the discrete-time price dynamics of a financial asset, when the hypothesis a special -period binomial model and the alternative is a different -period binomial model. As the observation gaps tend to zero (i. e. ), we obtain the limits of the corresponding Bayes risk as well as of the related Hellinger integrals and power divergences. Furthermore, we also give an example for the “non-commutativity” between Bayesian statistical and...