JOURNAL OF STATISTICAL RESEARCH OF IRAN (JSRI)
Year:2009 | Volume:6 | Issue:2|Papers Count:7

Principal components analysis is a well-known statistical method in dealing with large dependent data sets. It is also used in functional data for both purposes of data reduction as well as variation representation. On the other hand “handwriting” is one of the objects, studied in various statistical fields like pattern recognition and shape analysis. Considering time as the argument, the handwriting would be an infinite dimensional data; a functional object. In this paper we try to use thefunctional principal components analysis(FPCA) to the Persian handwriting data, analyzing the word Mehrwhich is the Persian term for Love.

A hybrid censoring scheme is a mixture of type I and type II censoring schemes. When n items are placed on a life test, the experiment terminates under type I or type II hybrid censoring scheme if either a prefixed censoring time T or the rth (1 £ r £ n is fixed) failure is first or later observed, respectively. In this paper, we investigate the decomposition of entropy in both hybrid censoring schemes. Entropy of type I hybrid censoring scheme is formulated and in order to determining the entropy of type II hybrid censoring scheme the available information are used. The results are then applied to some common life time distributions as illustrative examples.Finally, maximum entropy of the mentioned censoring schemes is discussed.

One of the factors which influence individual’s decision for presence or absence in the labor force is their personal features. In order to take appropriate policies to create employment and remove its obstacles in the countries labor market, we need to know the factors mentioned above and determine the amount and direction in which each factor influences the probability of individual’s presence in the labor force. The objective of this study is to determine how personal features influence persons’ decision on entering or leaving the labor market, during a one year period, for whole working age population, males and females. To achieve such objectives, using the panel data corresponding to spring 2005 and 2006 of the labor force survey conducted by the Statistical Centre of Iran, we distinguished the two subpopulation: active and inactive people in working age in 2005 and followed their economic activity status in 2006. To determine the effective factors on the change of individual’s activity status during a one-year period, we used a logit model and took advantage of the odds ratios to interpret the results of the model. The results of the study shows that substantial differences exist between men and women in terms of the stability of presence or absence in the labor force in a one-year period, however the amount of stability is also influenced by the individual features in both groups. The important finding of this paper is that not only the probability of inactive females entering the labor market is less than that for males, but also for women who are already in the labor market, the probability of leaving the market is more than that for men. However, the probability of returning to the labor market for women is also larger than that for men.

We study the distribution of discounted collective risk model where the counting process is Poisson. For the model considered here, we obtain mean, variance and moment generating function (m.g.f) of the model.To do this, we use two approaches. In the first approach we use classical methods to obtain the mean and variance. In the second approach we introduce some proper martingale and then we obtain the m.g.f of total loss by features of martingales. Additionally, we use Fast Fourier Transform to numerically calculate the distribution of discounted collective risk model.

In this paper, we derived recurrence relations for joint moment generating functions of nonadjacent generalized order statistics (GOS) of random samples drawn from doubly truncated class of continuous distributions.Recurrence relations for joint moments of nonadjacent GOS (ordinary order statistics (OOS) and k-upper records (k-RVs) as special cases) are obtained. Single and product moment generating functions (moments) of nonadjacent GOS are derived. Doubly truncated new modified Weibull (Weibull, Extreme-value, exponential and Rayleigh), three Burr type XII (Lomax) and inverse Weibull distributions, among others, arise as special cases of this doubly truncated class. Two applications are introduced, the first is the characterizations for members of the class based on recurrence relations for moments of GOS, OOS and k-RVs. As the second application we found Tables of single and product moments of OOS from doubly truncated Lomax distribution.

In this paper, we discuss a generalization of Balakrishnan skew normal distribution with two parameters that contains the skew-normal, the Balakrishnan skew-normal and the two-parameter generalized skew-normal distributions as special cases. Furthermore, we establish some useful properties and two extensions of this distribution.

Let {Xn, n³1} be a strictly stationary sequence of negatively associated random variables, with common distribution function F. In this paper, we consider the estimation of the two-dimensional distribution function of (X1, Xk+1) for fixed kÎN based on kernel type estimators. We introduce asymptotic normality and properties and moments. From these we derive the optimal bandwidth convergence rate, which is of order n-1.Besides of some usual conditions on the kernel function, the conditions typically impose a convenient increase rate on the covariancescov (X1, Xn).