Mohsen Maleki

Maximum a-posteriori estimation of autoregressive processes based on finite mixtures of scale-mixtures of skew-normal distributions

ABSTRACT This article investigates maximum a-posteriori (MAP) estimation of autoregressive model parameters when the innovations (errors) follow a finite mixture of distributions that, in turn, are scale-mixtures of skew-normal distributions (SMSN), an attractive and extremely flexible family of probabilistic distributions. The proposed model allows to fit different types of data which can be associated with different noise levels, and provides a robust modelling with great flexibility to accommodate skewness, heavy tails, multimodality and stationarity simultaneously. Also, the existence of convenient hierarchical representations of the SMSN random variables allows us to develop an EM-type algorithm to perform the MAP estimates. A comprehensive simulation study is then conducted to illustrate the superior performance of the proposed method. The new methodology is also applied to annual barley yields data.

Bayesian approach to epsilon-skew-normal family

ABSTRACT The estimation problem of epsilon-skew-normal (ESN) distribution parameters is considered within Bayesian approaches. This family of distributions contains the normal distribution, can be used for analyzing the asymmetric and near-normal data. Bayesian estimates under informative and non informative Jeffreys prior distributions are obtained and performances of ESN family and these estimates are shown via a simulation study. A real data set is also used to illustrate the ideas.

Two-Piece location-scale distributions based on scale mixtures of normal family

ABSTRACT In this work, we study the maximum likelihood (ML) estimation problem for the parameters of the two-piece (TP) distribution based on the scale mixtures of normal (SMN) distributions. This is a family of skewed distributions that also includes the scales mixtures of normal class, and is flexible enough for modeling symmetric and asymmetric data. The ML estimates of the proposed model parameters are obtained via an expectation-maximization (EM)-type algorithm.

The Skew-Reflected-Gompertz distribution for analyzing symmetric and asymmetric data

Abstract In this work, we have defined a new family of skew distribution: the Skew-Reflected-Gompertz. We have also derived some of its probabilistic and inferential properties. The maximum likelihood estimates of the proposed distribution parameters are obtained via an EM-algorithm, and performances of the proposed model and its estimates are shown via simulation studies as well as real applications. Three real datasets are also used to illustrate the model performance which can compete against some well-known skew distributions frequently used in applications.

Large Sample Inference about the Ratio of Means in Two Independent Populations

In a number of situations, for example in biology, psychology and neurosciences, researchers are interested in the ratio of two measured quantities. In this paper, we give an overview of different methods to constructing confidence limits for the ratios. Then by using the limiting theorems, a pivotal quantity for the ratio of population means will be presented. The results will be applied to construct confidence intervals and perform test of hypothesis. Finally, to investigate the ability of the proposed method, a modest simulation study is provided.

A Bayesian Approach to Robust Skewed Autoregressive Processes

Abstract This article studies autoregressive (AR) models assuming innovations with scale mixtures of skew-normal (SMSN) distributions, an attractive and flexible family of probability distributions. A Bayesian analysis considering informative prior distributions is presented. Comprehensive simulation studies are performed to support the performance of the proposed model and methods. The proposed methods are also applied on a real-time series data which has previously been analysed under Gaussian and Student-t AR models.

Time series models based on the unrestricted skew-normal process

ABSTRACT The standard location and scale unrestricted (or unified) skew-normal (SUN) family studied by Arellano-Valle and Genton [On fundamental skew distributions. J Multivar Anal. 2005;96:93–116] and Arellano-Valle and Azzalini [On the unification of families of skew-normal distributions. Scand J Stat. 2006;33:561–574], allows the modelling of data which is symmetrically or asymmetrically distributed. The family has a number of advantages suitable for the analysis of stochastic processes such as Auto-Regressive Moving-Average (ARMA) models, including being closed under linear combinations, being able to satisfy the consistency condition of Kolmogorov’s theorem and providing the guarantee of the existence of such a SUN stochastic process. The family is able to be represented in a hierarchical form which can be used for the ease of simulation. In addition, it facilitates an EM-type algorithm to estimate the model parameters. The performances and suitability of the proposed model are demonstrated on simulations and using two real data sets in applications.

Testing the equality of two independent regression models

ABSTRACT In some situations, for example in agriculture, biology, hydrology, and psychology, researchers wish to determine whether the relationship between response variable and predictor variables differs in two populations. In other words, we are interested in comparing two regression models for two independent datasets. In this work, we will use the parametric and nonparametric methods to establish hypothesis testing for the equality of two independent regression models. Then the simulation study is provided to investigate the performance of the proposed method.

A comparative analysis of classifiers in cancer prediction using multiple data mining techniques

In recent years, application of data mining methods in health industry has received increased attention from both health professionals and scholars. This paper presents a data mining framework for detecting breast cancer based on real data from one of the Iran hospitals by applying association rules and the most commonly used classifiers. The former were adopted for reducing the size of datasets, while the latter were chosen for cancer prediction. A k-fold cross-validation procedure was included for evaluating the performance of the proposed classifiers. Among the six classifiers used in this paper, support vector machine achieved the best results, with an accuracy of 93%. It is worth mentioning that the approach proposed can be applied for detecting other diseases as well.