Journal of the Iranian Statistical Society
http://jirss@irstat.ir
Journal of The Iranian Statistical Society - Journal articles for year 2017, Volume 16, Number 2Yektaweb Collection - http://www.yektaweb.comen2017/12/10On Concomitants of Order Statistics from Farlie-Gumbel-Morgenstern Bivariate Lomax Distribution and its Application in Estimation
http://jirss.irstat.ir/browse.php?a_id=303&sid=1&slc_lang=en
<p>In this paper, we have described the distribution theory of concomitants of order statistics arising from the Farlie-Gumbel-Morgenstern bivariate Lomax distribution. We have discussed the estimation of the parameters associated with the distribution of the variable Y of primary interest, based on ranked set sample defined by ordering the marginal observations on an auxiliary variable X, when (X, Y ) follows a Farlie-Gumbel-Morgenstern bivariate Lomax distribution. When the association parameter and the shape parameter corresponding to Y are known, we have proposed four estimators viz. an unbiased estimator based on Stokes' ranked set sample, the best linear unbiased estimator based on Stokes' ranked set sample, an unbiased estimator based on extreme ranked set sample and the best linear unbiased estimator based on multistage extreme ranked set sample for the scale parameter of the variable of primary interest. The relative efficiencies of these estimators have also been worked out.</p>
Anne PhilipA New Method For Generating Continuous Bivariate Families
http://jirss.irstat.ir/browse.php?a_id=289&sid=1&slc_lang=en
<p>The study presents a new method for generating continuous bivariate distributions. The random variables, X and Y are the transformers and bivariate random variable, T is the transformed. Based on the proposed method, many new bivariate distributions will be defined which can have ample usage in fitting the best distribution to data.</p>
Hossein BevraniSome theoretical results for two-step smoothing estimation of parameter from time-variant parametric models
http://jirss.irstat.ir/browse.php?a_id=348&sid=1&slc_lang=en
<p>In this article, we consider two nonparametric smoothing estimators for smoothing estimation of parameter. This parameter could be from any parametric family or from any parametric or semi-parametric regression model. Our estimation is based on two-step procedure, in which we first get the raw estimator of the parameter at a set of disjoint time points and then compute the final estimator at any time by smoothing the raw estimators. We derived asymptotic properties such as asymptotic biases, variances and mean squared errors (MSE) for the local polynomial smoothed estimator and kernel smoothing estimator. A mathematical relationship is established between two smoothing estimators. Mathematical relationship between two asymptotic MSEs has also been established.</p>
Mohammed ChowdhuryPrediction for Lindley Distribution Based on Type-II Right Censored Samples
http://jirss.irstat.ir/browse.php?a_id=358&sid=1&slc_lang=en
<p>Lindley distribution has received a considerable attention in the sta-<br>
tistical literature due to its simplicity. In this paper, we consider the<br>
problem of predicting the failure times of experimental units that are<br>
censored in a right-censored experiment when the underlying lifetime is<br>
Lindley distributed. The maximum likelihood predictor, the best unbiased<br>
predictor and the conditional median predictor are derived. Prediction in-<br>
tervals based on these predictors are considered. We further propose two<br>
resampling-based procedures for obtaining the prediction intervals. A nu-<br>
merical example is used to illustrate the methodology developed in this<br>
paper. Finally, a Monte Carlo simulation study is employed to evaluate<br>
the performance of diﬀerent prediction methods.</p>
Akbar AsgharzadehImproved estimation in Rayleigh type-II censored data under a bounded loss utilizing a point guess value
http://jirss.irstat.ir/browse.php?a_id=388&sid=1&slc_lang=en
<p><span class="fontstyle0">The problem of shrinkage testimation for the Rayleigh scale parameter </span><span class="fontstyle2">θ </span><span class="fontstyle0">based on censored samples under the reflected gamma loss function is considered. We obtain the minimum<br>
risk estimator and compute its risk. A shrinkage testimator based on acceptance or rejection<br>
of a null hypothesis </span><span class="fontstyle2">H</span><span class="fontstyle3">0 </span><span class="fontstyle0">: </span><span class="fontstyle2">θ </span><span class="fontstyle0">= </span><span class="fontstyle2">θ</span><span class="fontstyle3">0 </span><span class="fontstyle0">is constructed, where </span><span class="fontstyle2">θ</span><span class="fontstyle3">0 </span><span class="fontstyle0">is a point guess value of </span><span class="fontstyle2">θ</span><span class="fontstyle0">. The<br>
risk of the proposed shrinkage testimator is computed numerically and compared with the<br>
minimum risk estimator. One data set is used for illustrative purposes.</span><br style="line-height: normal; text-align: -webkit-auto; text-size-adjust: auto;" >
</p>
Mehran Naghizadeh QomiSome Results on Weighted Cumulative Entropy
http://jirss.irstat.ir/browse.php?a_id=380&sid=1&slc_lang=en
<p>Considering Rao et al. (2004) and Di Crescenzo and Longobardi (2009) studies, we deﬁne a new measure of information based on cumulative distribution function, which is called “weighted cumulative entropy” (WCE). The above-mentioned model is a “Shift-dependent uncertainty measure”. In addition, we examine some of the properties of WCE and obtain some bounds for this measure. In order to estimate this new information measure, we put forward empirical WCE. Furthermore, in some theorems we have some characterization results. We explore that if the ﬁrst (last) order statistics are equal for two distributions, then this two distributions will be equal.<br>
</p>
Simin BaratpourSkew Laplace Finite Mixture modelling
http://jirss.irstat.ir/browse.php?a_id=409&sid=1&slc_lang=en
<p></p>
<div dir="ltr" font-size:="" id="m_-892245609597034382yui_3_16_0_ym19_1_1492992463992_5365" new="" style="color: rgb(0, 0, 0); font-family: " times=""><span id="m_-892245609597034382yui_3_16_0_ym19_1_1492992463992_6175">This paper presents a new mixture model via considering the univariate skew Laplace </span>distribution. The new model can handle both heavy tails and skewness and is multimodal. Describing some properties of the proposed model, we present a feasible EM algorithm for iteratively computing maximum likelihood estimates. We also derive the observed information matrix for obtaining the asymptotic standard errors of parameter estimates. The finite sample properties of the obtained estimators together with the consistency of the associated standard error of parameter estimates are investigated by a simulation study. We also demonstrate the <span id="m_-892245609597034382yui_3_16_0_ym19_1_1492992463992_6226">flexibility and usefulness of the new model by analyzing real data example.</span> </div>
Mehrdad Naderi