Recursos de colección
Project Euclid (Hosted at Cornell University Library) (192.320 recursos)
Afrika Statistika
Afrika Statistika
ELBATAL, Ibrahim; EL GEBALY, Yehia Mousa; AMIN, Essam Ali
This paper introduces a new five-parameter lifetime model called the exponentiated Kumaraswamy power Lindley distribution, that extends the power Lindley distribution and some well-known distributions. Various structural properties of the new model including expansions for the density function, explicit expressions for the ordinary and conditional moments, residual and reversed residual life functions and mean deviations are derived. The maximum likelihood method is used to estimate the model parameters. The usefulness and flexibility of the proposed model are illustrated empirically by means of two real data sets.
MOKRANI, Fatiha; FELLAG, Hocine; NECIR, Abdelhakim
In this work, robust Bayesian estimation of the generalized Pareto distribution is proposed. The methodology is presented in terms of oscillation of posterior risks of the Bayesian estimators. By using a Monte Carlo simulation study, we show that, under a suitable generalized loss function, we can obtain a robust Bayesian estimator of the model.
ANANI, Lotsi; WIT, Ernst
This paper considers the problem of networks reconstruction from heterogeneous data using a Gaussian Graphical Mixture Model (GGMM). It is well known that parameter estimation in this context is challenging due to large numbers of variables coupled with the degenerate nature of the likelihood. We propose as a solution a penalized maximum likelihood technique by imposing an l1 penalty on the precision matrix. Our approach shrinks the parameters thereby resulting in better identifiability and variable selection.
SECK, Cheikh Tidiane; DIAM, Ba; LO, Gane Samb
We propose nonparametric asymptotic confidence intervals for the upper and lower tail dependence coefficients. These latter are obtained from confidence bands established for the copula function itself and based upon three kernel-type estimators. We show the performance of these confidence intervals through a simulation study. We also apply these results to financial data stemming from the CAC 40 stock index which reveals the existence of extreme dependence between larger values of the opening and closing prices for this index during the considered period.
MOKKADEM, Abdelkader; PELLETIER, Mariane
The aim of this paper is to provide pointwise and uniform moderate deviations principles for the kernel
estimator of a nonrandom regression function. Moreover, we give an
application of these moderate deviations principles to the construction of confidence regions
for the regression function.
BERECHE, Aicha; CHERFAOUI, Mouloud; AÏSSANI, Djamil
This paper aims to establish the effect of the choice of a stability bound for the ruin probability on the quality of the approximation of the characteristics (ruin probabilities) of two classical risk models to approach (ideal an perturbed models) regarding to different large claims. In particular, we use two versions of the strong stability method: strong stability of a Markov chain and strong stability of a Lindley process. A comparative study, based on numerical results obtained by simulation, is performed between the two versions.
CISS, Youssou; DIAKHABY, Aboubakary
In this paper, we study the kernel estimator of Foster, Greer and Thorbecke class of measures when the poverty aversion parameter is strictly between zero and one, as a genreralization of the work of Dia(2016). We solved an open problem arising in mentioned paper. The asymptotic normality of the estimator is established. As an illustration, we determine the confidence intervals for different regions of Senegal. The study of this application demonstrated that our methodology is not only more efficient than the empirical estimator, but it also provides better confidence intervals for the poverty index.
LARBI, Lydia; FELLAG, Hocine
In this work, robust Bayesian analysis of the Bayesian estimation of an autoregressive model with exponential innovations is
performed. Using a Bayesian robustness methodology, we show that, using a suitable generalized quadratic loss, we obtain optimal
Bayesian estimators of the parameters corresponding to the smallest oscillation of the posterior risks.
MAKINDE, Olusola Samuel
We study the theoretical misclassification probability of linear and quadratic classifiers and examine the performance of these
classifiers under distributional variations in theory and using simulation. We derive expression for Bayes errors for some competing
distributions from the same family under location shift.
Benelmir, Imane; Meraghni, Djamel
We apply the inversion method of estimation, with several combinations of two
among the four most popular association measures, to estimate the parameters of copulas in the case of bivariate distributions.
We carry out a simulation study with two examples, namely
Farlie-Gumbel-Morgenstern and Marshall-Olkin two-parameter copulas to
make comparisons between the obtained estimators, with respect to bias and root of the
mean squared error.
ZEGHDOUDI, Halim; NEDJAR, Sihem
In this paper, we introduce a new distribution named as the Pseudo Lindley Distribution (PsLD) as a generalization of the
Lindley distribution (LD). A full and detailed description are provided in terms of moments, cumulates, characteristic function,
failure, rate function, stochastic ordering, distributions of sums, and parameters estimation. Simulations studies and data driven
applications are also reported.
CISSE, Papa Ousmane; DIONGUE, Abdou Kâ; GUEGAN, Dominique
In this paper we introduce a new model called Fractionally Integrated Separable Spatial Autoregressive processes with
Seasonality and denoted Seasonal FISSAR. We focus on the class of separable spatial models whose correlation structure can be
expressed as a product of correlations. This new modelling allows taking into account the seasonality patterns observed in spatial
data. We investigate the properties of this new model providing stationary conditions, some explicit form of the autocovariance
function and the spectral density. We also establish the asymptotic behaviour of the spectral density function near the seasonal
frequencies.
LOUIZA, Soltane; DJAMEL, Meraghni; NECIR, Abdelhakim
Many insurance premium principles are defined and
various estimation procedures introduced in the literature.
In this paper, we focus on the estimation of the
excess-of-loss reinsurance premium when losses are randomly
censored. The asymptotic normality of the proposed estimator
is established under suitable conditions and its performance
evaluated through sets of simulated data.
BRAHIMI, Brahim; ZOUBIR, Kenioua
We use the so-called t-Hill tail index estimator
proposed by Fabián (2001), rather than Hill's one, to derive
a robust estimator for the distortion risk premium of
losses. Under the second-order condition of regular
variation, we establish its asymptotic normality. By
simulation study, we show that this new estimator is more
robust than of Necir and Meraghni (2009) both for small and
large samples.
GANE SAMB, LO; HAROUNA, Sangare; NDIAYE, Cheikhna Hamallah
Association between random variables is a
generalization of independence of these random variables.
This concept is more and more commonly used in current
trends in any research fields in Statistics. In this paper,
we proceed to a simple, clear and rigorous introduction to
it. We will present the fundamental asymptotic normality
theorem on stationary and associated sequences of random
variables. A coherent and modern frame is used. This review
will be profitable to new researchers in the topic.
EWEMOOJE, Olusegun S.; AMAHIA, Godwin N.
We proposed new and more efficient estimators for estimating population
proportion of respondents belonging to two related sensitive attributes in survey sampling by
extending the work of Mangat(1994). Our proposed estimators are more efficient than Lee et al.
(2013) simple and crossed model estimators as the population proportion of possessing the
sensitive attribute increases.
NZABANIT, Joseph; VON ROSEN, Dietrich; SINGULL, Martin
In this paper, the bilinear regression model based on normally distributed random matrix is studied.
For these models, the dispersion matrix has the so called Kronecker product structure and
they can be used for example to model data with spatio-temporal relationships.
The aim is to estimate the parameters of the model when, in addition, one covariance
matrix is assumed to be linearly structured. On the basis of n independent observations
from a matrix normal distribution, estimating equations in a flip-flop relation are
established and the consistency of estimators is studied.
AHSANULLAH, Mohammad
Rider(1957) introduced the generalized Cauchy distribution. In this paper we
will consider a truncated version of the generalized Cauchy distribution. One possible
use of the model is in life testing problem when the domain of application is non
negative. Here we will consider several distributional properties of the truncated
generalized Cauchy distribution. Someome characterizations of this distribution based on
truncated moments, order statistics and record values, will be given.
Lessak, Radia; Mohdeb, Zaher
In this paper, we develop a test for the nonparametric regression model in the case of a homoscedastic error structure and fixed design.
More precisely, a new statistic is proposed for testing linear hypothesis versus regime switching alternatives; without regularity condition,
and also under either the null or the alternative hypotheses. We establish the asymptotic normality of the test statistic under the null hypothesis
and the alternative one. A simulation study is conducted to investigate the finite sample properties of the proposed test.
Bennour, Besma; Belaloui, Soheir
A consecutive k-out-of-n system consists of an ordered sequence of $n$ components, such that the system functions if and only if at
least $k$ $(k\leq n)$ consecutive components function. The system is called linear ($L$) or circular ($C$) depending on whether the components are
arranged on a straight line or form a circle. In the first part, we use a shock model to obtain the reliability function of consecutve-k-out-of-n
systems with dependent and nonidentical components. In the second part, we treat some numerical examples to show the derive results and deduce the
failure rate of each component and the system.