Most recent empirical work implies that the presence of low-dimensional deterministic chaos increases the complexity of the financial time series behavior. In this study we propose the Generalized Multilayer Perceptron (GMLP), and the Bayesian inference via Markov Chain Monte Carlo (MCMC) method for parameter estimation and one-step-ahead prediction. By out-of-sample prediction approach, these proposed methods are compared to autoregressive integrated moving average (ARIMA) models which have been used as a benchmark. The deterministic Mackey-Glass equation with errors that follow an ARCH (p) process (MG-ARCH (p)) is applied to generate the data set used in this study. It turns out that GMLP outperforms the other two forecasting methods using RMSE, MAPE, and MAE criteria of forecasting accuracy.

this paper we consider three parameter generalized exponential distribution. Exact expressions and some recurrence relations for single and product moments of lower generalized order statistics are derived. Further the results are deduced for moments of order statistics and lower records and characterization of this distribution has been considered on using the conditional moment of the lower generalized order statistics.

A generalized version of four parameter modified inverse weibull distribution (MIWD) is introduced in this paper. This distribution generalizes the following distributions: (1) Modified Inverse exponential distribution, (2) Modified Inverse Rayleigh distribution, (3) Inverse weibull distribution. We provide a comprehensive description of the mathematical properties of the modified inverse weibull distribution along with its reliability behaviour. We derive the moments, moment generating function and examine the order statistics. We propose the method of maximum likelihood for estimating the model parameters and obtain the observed information matrix.

In this paper we propose a new asymptotically normal estimator of the reinsurance premium for the losses distribution. Our estimator is based on the reduced bias of the extreme quantile and the index of an heavy-tailed distribution. Moreover, we illustrate the behaviour of the proposed estimator and give a comparison between this estimator and the classical semi parametric estimator proposed by Necir et al. (2007) in terms of the bias and the root mean squared error (rmse).

A method that makes use of combinatorics for selecting N objects out of distinguishable objects is developed for constructing D-optimal N-point exact designs. The difficulties which are experienced in the variance exchange algorithms for constructing D-optimal exact designs, such as cycling, slow convergence and failure to converge to the desired optimum, are not experienced by this method. The method converges rapidly and absolutely to the desired N-point D-optimal design and is effective for determining optimal designs in block experiments as well as in non-block experiments for finite or infinite number of support points in the space of trials.

In this paper, taking into account the possible development of serious disorders of the proliferation of the plasmatic cells, we focus on a dataset concerning the prediction among a chronic disease which has the higher risk of malignant transformation. The purpose of this paper is to argue in favour of the use of multiple correspondence analysis (MCA) as a powerful exploratory tool for such data. Following usual regression terminology, we refer to the primary variable as the response variable and the others as explanatory or predictive variables. As an alternative, a copula based methodology for prediction modeling and an algorithm to stimulate data are proposed.