Nursel Koyuncu

On improvement in estimating population mean in stratified random sampling

Gupta and Shabbir 2 have suggested an alternative form of ratio-type estimators for estimating the population mean. In this paper, we obtained a corrected version for the mean square error (MSE) of the Gupta–Shabbir estimator, up to first order of approximation, and the optimum case is discussed. We expand this estimator to the stratified random sampling and propose general classes for combined and separate estimators. Also an empirical study is carried out to show the properties of the proposed estimators.

Family of Estimators of Population Mean Using Two Auxiliary Variables in Stratified Random Sampling

A general family of estimators, which use the information of two auxiliary variables in the stratified random sampling, is proposed to estimate the population mean of the variable under study. Under stratified random sampling without replacement scheme, the expressions of bias and mean square error (MSE) up to the first- and second-order approximations are derived. The family of estimators in its optimum case is discussed. Also, an empirical study is carried out to show the properties of the proposed estimators.

Efficient estimators of population mean using auxiliary attributes

Abstract Abd-Elfattah et al. [1] suggested a set of estimators for calculating population mean using auxiliary attributes. This paper proposes a family of estimators based on an adaptation of the estimators presented by Koyuncu and Kadilar [2] , and introduces a new family of exponential estimators using auxiliary attributes. The expressions of the mean square errors (MSEs) of the adapted and proposed families are derived in a general form. It is shown that the adapted version of the Koyuncu and Kadilar [2] estimators is always more efficient than that of Abd-Elfattah et al. [1] . Moreover, the new exponential estimators based on auxiliary attributes are more efficient than those of Koyuncu and Kadilar [2] and Abd-Elfattah et al. [1] . The theoretical findings are supported by a numerical example using original data.

Calibration Weighting in Stratified Random Sampling

A new calibration estimator is proposed to estimate the population mean in the stratified random sampling. The corrected expression of Tracy et al. (2003) calibrated weights are presented and new improved calibration weights are introduced. Theoretical variance of the suggested estimator is discussed. Also a simulation study is carried out to show the properties of the proposed estimator.

New ranked set sampling schemes for range charts limits under bivariate skewed distributions

In this paper, instead of using traditional sampling schemes, we newly introduce to use ranked set, neoteric ranked set and median ranked set sampling to construct range charts control limits under the bivariate skewed distributions. Shewhart, weighted variance and skewness correction methods are considered to construct the range charts in the statistical process control. We have compared the performance of the suggested range charts under simple random and different ranked set sampling schemes by simulation study. The simulation study has showed that range charts based on skewness correction method are the most efficient under neoteric and median ranked set sampling schemes, especially in highly skewed data.

New difference-cum-ratio and exponential type estimators in median ranked set sampling

This paper suggests difference-cum-ratio and exponential type estimators of population mean using first or third quartiles and mean of auxiliary variable under median ranked set sampling scheme and we have extended our study in double sampling scheme when the population parameters are unknown. The bias and mean square error of estimators are derived by theoretically both of sampling designs. Empirical studies have been done to demonstrate the efficiency of proposed estimators over the existing estimators. We have found that difference-cum-ratio estimator is always more efficient than regression estimator and both of the estimators are considerable efficient than existing estimators.

On efficiency of cluster sampling on sampling on two occasions when there is non-response

Abstract In this article we endeavor the issue of estimation of the entirety of mean and the difference in mean in mail reviews. This issue is directed for current occasion with regards to group examining (cluster sampling) over inspecting on two progressive occasions when there is non-response (i) on one and two (both) occasions, (ii) just on the first occasion and (iii) just on the second occasion. An examination between variances of the estimates is discussed. The outcomes got are exhibited with the assistance of simulation study.

Calibration estimator of population mean under stratified ranked set sampling design

ABSTRACT Calibration method adjusts the original design weights to improve the estimates by using auxiliary information. In this article we have proposed new calibration estimators under stratified ranked set sampling design and derive the estimator of variance of calibration estimator. A simulation study is carried out to see the performance of proposed estimators.

Preface of the "Symposium on advances in sampling theory and applications"

The mini symposium “Advances in Sampling Theory and Applications” will be arranged in the conference “11th International Conference of Numerical Analysis and Applied Mathematics” during 21-27 September 2013 in Rhodes. The aim of this meeting is to bring together mathematicians and statisticians interested in sampling theory and its applications to related field store view recent advances and discuss the problems. Scientific researchers have submitted several papers related developments within traditional sampling theory, modern techniques, randomized response models, stochastic process, time series etc. Their studies are evaluated for oral presentation and poster presentation. There will be nine talks and one poster presentation. Emrah Altun will give a talk “A Comparison of Portfolio Selection Models via Application on ISE 100 Index Data”. In this study he compared different models used in finance. Hatice Oncel Cekim will give a talk “Characterizing Forest Fire Activity in Turkey by Compound Poisson and Time Series Models”. She will introduce compound poisson model and time series model using the forest fires data occurred in Turkey. Selen Cakmakyapan will give a talk “On The Expected Value and Distribution Function of the First Exit Time for the Polya-Aeppli Process”. In her study the distribution function and expected value of the first exit time are derived for Polya-Aeppli process. She used traffic accidents in Groningen in her study using R Project. Miroslav Voˇrechovský will talk “Correlation Process Resulting from Random Swapping of the Order of Elements”. In his study he deals with the properties of a process of statistical correlation between two vectors of samples of random variables. Semra Turkan will give a talk “Examination of Influential Observations in Penalized Spline Regression”. She adapted Pena’s approach to penalized spline regression in terms of ordinary residuals and leverages. The real data and artificial data are used to see illustrate the effectiveness of Pena’s statistic as to Cook’s distance on detecting influential observations. Esra Polat will give a talk “The Comparison of Robust Partial Least Squares Regression with Robust Principal Component Regression on a Real Data”. The purpose of her study is to show the usage of robust methods on an econometric data set and making a comparison of two methods on an inflation model of Turkey. Gamze Ozel Kadilar will give a talk “Lindley Frailty Model for a Class of Compound Poisson Processes”. She will introduce frailty models with discrete compound Poisson processs for the Lindley distributed failure time. More over survival functions are derived and maximum likelihood estimation procedures for the parameters are studied. Then she illustrates her results using earthquake data set of Turkey. Nursel Koyuncu will give a talk “Efficient Combined Estimators of Population Mean using Auxiliary Attribute under Stratified Random Sampling”. She deals with the problem of estimating the finite population mean when information on auxiliary attribute is available under stratified random sampling scheme. New exponential type estimators are introduced. Nilgun Ozgul will give a talk “A New Mean Estimator using Auxiliary Variables for Randomized Response Models”. She suggested a new mean estimator for RRM including quantitative responses and derived the meansquare error. To show the efficiency of the proposed estimator to other existing estimators in RRM she performed a simulation study.