In risk analysis based on Bayesian framework, premium calculation requires specification of a prior distribution for the risk parameter in the heterogeneous portfolio. When the prior knowledge is vague, the E-Bayesian and robust Bayesian analysis can be used to handle the uncertainty in specifying the prior distribution by considering a class of priors instead of a single prior. In th...

In this paper, the well-known proportional hazards model which includes several well-known lifetime distributions such as exponential,Pareto, Lomax, Burr type XII, and so on is considered. With both Bayesian and non-Bayesian approaches , we consider the estimation of parameters of interest based on progressively Type-II right censored samples. The Bayes estimates are obtained based on symmetric...

in a finite stationary markov chain, transition probabilities may depend on some explanatoryvariables. a similar problem has been considered here. the corresponding posteriors are derived andinferences are done using these posteriors. finally, the procedure is illustrated with a real example.

Genetic programming (GP) can automatically fuse given classifiers of diverse types to produce a combined classifier whose Receiver Operating Characteristics (ROC) are better than [Scott et al.1998b]’s “Maximum Realisable Receiver Operating Characteristics” (MRROC). I.e. better than their convex hull. This is demonstrated on a satellite image processing bench mark using Naive Bayes, Decision Tre...

In this paper classifiers for text-based prediction of intonation contour classes are compared. The contour classes were derived automatically by a method presented in Reichel (2006), and the following classifiers were utilised for prediction: Bayes classifier, C4.5 decision trees, perceptrons, and linear feedforward networks. Prediction accuracies amounted from 38.0% (Perceptron) to 66.6% (Lin...

With a goal of understanding what drives generalization in deep networks, we consider several recently suggested explanations, including norm-based control, sharpness and robustness. We study how these measures can ensure generalization, highlighting the importance of scale normalization, and making a connection between sharpness and PAC-Bayes theory. We then investigate how well the measures e...

*Abstract—The main purpose of this project is to analyze several Machine Learning techniques individually and compare the efficiency and classification accuracy of those techniques. Three algorithms are used (Naïve Bayes learning, feed forward Artificial Neural Networks with Backpropagation, and Decision Trees learning using C4.5) over two datasets (“European companies” and “Japanese companies”...

When modelling a probability distribution with a Bayesian network, we are faced with the problem of how to handle continuous variables. Most previous works have solved the problem by discretizing them with the consequent loss of information. Another common alternative assumes that the data are generated by a Gaussian distribution (parametric approach), such as conditional Gaussian networks, wit...