Omer Ozturk
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Omer Ozturk has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Olena Kravchuk
School of Agriculture, Food, and Wine The University of Adelaide Adelaide South Australia Australia
10 shared publications Douglas A. Wolfe
Department of Statistics , The Ohio State University , 1958 Neil Avenue, Columbus, OH, 43210, USA
6 shared publications- Jennifer Ann Brown
University of Canterbury
3 shared publications Narayanaswamy Balakrishnan
Department of Mathematics and Statistics, MacMaster University, Ontario, Canada
3 shared publicationsJinguo Gao
Alphabet Inc., Mountain View, CA, USA
2 shared publicationsOlena Kravchuk
School of agriculture, Food and Wine, University of Adelaide, Adelaide, Australia
2 shared publications
Latest Publications
- Open Access
- 28 October 2026
- Published by Wiley in Canadian Journal of Statistics
- p. e70077
Order restricted row‐column cluster randomized design
This research introduces a novel two‐stage cluster randomized design, the order restricted row‐column cluster randomized design. The new design builds upon the cluster randomized block design by incorporating a second layer of blocking, achieved through ranking cluster units that are randomly sampled from the population. This approach creates a two‐way layout, with blocks and ranking groups, and employs restricted randomization to enhance the accuracy of treatment contrast estimation. We calculate the expected mean square for each source of variation in this new design under a suitable linear model, develop an approximate ‐test for the treatment effect, assess the impact of ranking quality, calculate optimal sample sizes for a given cost model, formulate multiple comparison procedures, and apply the design to an educational setting.
- 28 October 2026
- Published by Wiley in Canadian Journal of Statistics
- p. e70077
Open AccessOrder restricted row‐column cluster randomized design
This research introduces a novel two‐stage cluster randomized design, the order restricted row‐column cluster randomized design. The new design builds upon the cluster randomized block design by incorporating a second layer of blocking, achieved through ranking cluster units that are randomly sampled from the population. This approach creates a two‐way layout, with blocks and ranking groups, and employs restricted randomization to enhance the accuracy of treatment contrast estimation. We calculate the expected mean square for each source of variation in this new design under a suitable linear model, develop an approximate ‐test for the treatment effect, assess the impact of ranking quality, calculate optimal sample sizes for a given cost model, formulate multiple comparison procedures, and apply the design to an educational setting.
- Open Access
- 18 March 2026
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
Augmented Spatially Balanced and Probability Proportional-to-Size Samples
In this paper, we examine a finite population distributed across a two-dimensional spatial domain, where a small subset of units exhibit substantially larger response values compared to the rest. Each unit is associated with a size variable, which is proportional to its response value and known prior to sampling. From this population, we develop an augmented sampling strategy involving two randomization stages: a spatially balanced sample ( SBS ) and a probability proportional-to-size (PPS) sample with unequal inclusion probabilities. First, we select an SBS of size $$n_1$$ , ensuring spatial balance across the population. Subsequently, a local pivotal method ( LPM ) sample of size $$n_2$$ is drawn from the remaining unsampled units, using unequal inclusion probabilities proportional to the size variable. This two-stage sampling approach combines spatial balance with targeted selection of large-size units. Our results show that the population mean estimator derived from the proposed sampling design outperforms those based on individual SBS or LPM samples. Furthermore, we provide explicit expressions for the first-order inclusion probabilities associated with a given size variable. To illustrate the method’s effectiveness, we apply the proposed design to a population of poppy fields in Kandahar province of Afghanistan. Supplementary materials accompanying this paper appear online.- 18 March 2026
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
Open AccessAugmented Spatially Balanced and Probability Proportional-to-Size Samples
In this paper, we examine a finite population distributed across a two-dimensional spatial domain, where a small subset of units exhibit substantially larger response values compared to the rest. Each unit is associated with a size variable, which is proportional to its response value and known prior to sampling. From this population, we develop an augmented sampling strategy involving two randomization stages: a spatially balanced sample ( SBS ) and a probability proportional-to-size (PPS) sample with unequal inclusion probabilities. First, we select an SBS of size $$n_1$$ , ensuring spatial balance across the population. Subsequently, a local pivotal method ( LPM ) sample of size $$n_2$$ is drawn from the remaining unsampled units, using unequal inclusion probabilities proportional to the size variable. This two-stage sampling approach combines spatial balance with targeted selection of large-size units. Our results show that the population mean estimator derived from the proposed sampling design outperforms those based on individual SBS or LPM samples. Furthermore, we provide explicit expressions for the first-order inclusion probabilities associated with a given size variable. To illustrate the method’s effectiveness, we apply the proposed design to a population of poppy fields in Kandahar province of Afghanistan. Supplementary materials accompanying this paper appear online. - Open Access
- 12 October 2025
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
- p. 1-27
Post-Stratified Inference for Pairwise Treatment Comparisons in Randomized Block Designs
In this study, we introduce a novel approach to statistical inference for multiple comparison procedures in a generalized randomized complete block design (RCBD) by employing post-stratification of experimental units. After completing the experiment, our method involves randomly pairing within-block experimental units (EUs) subjected to two different treatments, t and $$t'$$ . Each pair of experimental units is then ranked based on pre-treatment auxiliary (covariate) information, assuming no treatment effects. These ranked sets are subsequently divided into two distinct categories, which serve as an additional blocking factor based on the ranking of within-set experimental units. The first ranking block contains the sets where the lower rank corresponds to treatment t and the higher rank to treatment $$t'$$ , while the second ranking block contains the reverse treatment allocation. This post-stratification is performed for all possible pairs of treatments with $$t<t'$$ . Based on this post-stratified data, we develop a multiple comparison procedure for all pairwise contrast parameters. The proposed procedure demonstrates superior statistical power and yields narrower confidence intervals for multiple pairwise contrast inferences when the assumptions of the analysis of covariance model are not satisfied. The application of the proposed procedure is illustrated using experimental data.- 12 October 2025
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
- p. 1-27
Open AccessPost-Stratified Inference for Pairwise Treatment Comparisons in Randomized Block Designs
In this study, we introduce a novel approach to statistical inference for multiple comparison procedures in a generalized randomized complete block design (RCBD) by employing post-stratification of experimental units. After completing the experiment, our method involves randomly pairing within-block experimental units (EUs) subjected to two different treatments, t and $$t'$$ . Each pair of experimental units is then ranked based on pre-treatment auxiliary (covariate) information, assuming no treatment effects. These ranked sets are subsequently divided into two distinct categories, which serve as an additional blocking factor based on the ranking of within-set experimental units. The first ranking block contains the sets where the lower rank corresponds to treatment t and the higher rank to treatment $$t'$$ , while the second ranking block contains the reverse treatment allocation. This post-stratification is performed for all possible pairs of treatments with $$t<t'$$ . Based on this post-stratified data, we develop a multiple comparison procedure for all pairwise contrast parameters. The proposed procedure demonstrates superior statistical power and yields narrower confidence intervals for multiple pairwise contrast inferences when the assumptions of the analysis of covariance model are not satisfied. The application of the proposed procedure is illustrated using experimental data. - Open Access
- 19 December 2023
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
- Vol. 29 (4) , 831-852
Order Restricted Randomized Block Designs
Field experiments are run under two competing objectives, high precision and minimal cost. The precision can be increased either by using sound experimentation techniques that account for the sources of variation with reasonable statistical models or by increasing the sample size. Large sample sizes usually increase the cost of the experiment and may not be feasible. This paper uses order restricted randomized designs (ORRD) to increase the precision while keeping the sample size and cost of the experiment minimal. The ORRD described here starts with a randomized block design but adds a second layer of blocking by ranking plots within each block. This creates a two-way lay-out, blocks and ranking groups, and uses a restricted randomization to improve the precision of estimation of the treatment parameters. Ranking groups create a correlation structure for within-block units. The restricted randomization uses this correlation structure to reduce the error variance of the experiment. The paper computes the expected mean square for each source of variation in the ORRD design under a suitable linear model. It also provides approximate F-tests for treatment and ranking group effects. The efficiency of the ORRD is investigated through empirical power studies. Finally, an example based on a uniformity field trial illustrates the use of the method in a split-plot experiment. Supplementary material to this paper is provided online.
- 19 December 2023
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
- Vol. 29 (4) , 831-852
Open AccessOrder Restricted Randomized Block Designs
Field experiments are run under two competing objectives, high precision and minimal cost. The precision can be increased either by using sound experimentation techniques that account for the sources of variation with reasonable statistical models or by increasing the sample size. Large sample sizes usually increase the cost of the experiment and may not be feasible. This paper uses order restricted randomized designs (ORRD) to increase the precision while keeping the sample size and cost of the experiment minimal. The ORRD described here starts with a randomized block design but adds a second layer of blocking by ranking plots within each block. This creates a two-way lay-out, blocks and ranking groups, and uses a restricted randomization to improve the precision of estimation of the treatment parameters. Ranking groups create a correlation structure for within-block units. The restricted randomization uses this correlation structure to reduce the error variance of the experiment. The paper computes the expected mean square for each source of variation in the ORRD design under a suitable linear model. It also provides approximate F-tests for treatment and ranking group effects. The efficiency of the ORRD is investigated through empirical power studies. Finally, an example based on a uniformity field trial illustrates the use of the method in a split-plot experiment. Supplementary material to this paper is provided online.
- 29 November 2023
- Published by Springer Nature in Environmental and Ecological Statistics
- Vol. 30 (4) , 741-767
Trade-off between efficiency and variance estimation of spatially balanced augmented samples
- 29 November 2023
- Published by Springer Nature in Environmental and Ecological Statistics
- Vol. 30 (4) , 741-767
Trade-off between efficiency and variance estimation of spatially balanced augmented samples
- Open Access
- 2 May 2023
- Published by Wiley in Statistics in Medicine
- Vol. 42 (17) , 2887-2913
Robust rank‐based meta‐analyses for two‐sample designs with application to platelet counts of malaria infection data
In this paper, we propose robust meta‐analysis procedures for individual studies that report a broad range of robust summary statistics for a two‐sample problem. Summary statistics of individual studies could be presented in different forms including full data, medians of the two samples, the Hodges‐Lehman and Wilcoxon estimates of the location shift parameters. Data synthesis is made under both fixed‐effect and random‐effect meta‐analysis models. We systematically compare these robust meta‐analysis procedures via simulation studies to meta‐analysis procedure based on sample means and variances from individual studies under a wide range of error distributions. We show that the coverage probabilities of the robust meta‐analysis confidence intervals are quite close to the nominal confidence level. We also show that mean square error (MSE) of the robust meta‐analysis estimator is considerably smaller than that of the non‐robust meta‐analysis estimator under the contaminated normal, heavy tailed and skewed error distributions. The robust meta‐analysis procedures are then applied to platelet count reduction for malaria infected patients in Ghana.
- 2 May 2023
- Published by Wiley in Statistics in Medicine
- Vol. 42 (17) , 2887-2913
Open AccessRobust rank‐based meta‐analyses for two‐sample designs with application to platelet counts of malaria infection data
In this paper, we propose robust meta‐analysis procedures for individual studies that report a broad range of robust summary statistics for a two‐sample problem. Summary statistics of individual studies could be presented in different forms including full data, medians of the two samples, the Hodges‐Lehman and Wilcoxon estimates of the location shift parameters. Data synthesis is made under both fixed‐effect and random‐effect meta‐analysis models. We systematically compare these robust meta‐analysis procedures via simulation studies to meta‐analysis procedure based on sample means and variances from individual studies under a wide range of error distributions. We show that the coverage probabilities of the robust meta‐analysis confidence intervals are quite close to the nominal confidence level. We also show that mean square error (MSE) of the robust meta‐analysis estimator is considerably smaller than that of the non‐robust meta‐analysis estimator under the contaminated normal, heavy tailed and skewed error distributions. The robust meta‐analysis procedures are then applied to platelet count reduction for malaria infected patients in Ghana.
- Open Access
- 11 April 2023
- Published by Wiley in Statistics in Medicine
- Vol. 42 (15) , 2692-2710
Models for cluster randomized designs using ranked set sampling
Cluster randomized designs (CRD) provide a rigorous development for randomization principles for studies where treatments are allocated to cluster units rather than the individual subjects within clusters. It is known that CRDs are less efficient than completely randomized designs since the randomization of treatment allocation is applied to the cluster units. To mitigate this problem, we embed a ranked set sampling design from survey sampling studies into CRD for the selection of both cluster and subsampling units. We show that ranking groups in ranked set sampling act like a covariate, reduce the expected mean squared cluster error, and increase the precision of the sampling design. We provide an optimality result to determine the sample sizes at cluster and sub‐sample level. We apply the proposed sampling design to a dental study on human tooth size, and to a longitudinal study from an education intervention program.
- 11 April 2023
- Published by Wiley in Statistics in Medicine
- Vol. 42 (15) , 2692-2710
Open AccessModels for cluster randomized designs using ranked set sampling
Cluster randomized designs (CRD) provide a rigorous development for randomization principles for studies where treatments are allocated to cluster units rather than the individual subjects within clusters. It is known that CRDs are less efficient than completely randomized designs since the randomization of treatment allocation is applied to the cluster units. To mitigate this problem, we embed a ranked set sampling design from survey sampling studies into CRD for the selection of both cluster and subsampling units. We show that ranking groups in ranked set sampling act like a covariate, reduce the expected mean squared cluster error, and increase the precision of the sampling design. We provide an optimality result to determine the sample sizes at cluster and sub‐sample level. We apply the proposed sampling design to a dental study on human tooth size, and to a longitudinal study from an education intervention program.
- Open Access
- 23 August 2022
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
- Vol. 27 (4) , 652-673
Row–Column Sampling Design Using Auxiliary Ranking Variables
In this paper, we present a new class of row–column sampling design when a sampling region is presented by non-overlapping quadrats of an rp-by-p row–column grid. Quadrats are rank-ordered on a relevant auxiliary variable, first in the rp rows of the grid and then in the so formed p ranked columns. This ranking enhances the precision of the sampling using ideas from standard rank set sampling methods. The sampling design falls in the traditional sampling framework where the sample selection probabilities are independent of the variables of interest. Selection probabilities are governed by a spatial design and the auxiliary information to facilitate better sampling coverage over the sampling region. The paper constructs unbiased estimators for population mean, total and provides variance estimates for them. It also shows when the new sampling design performs better than its competitors. The new sampling design is applied to a forest and two agricultural field samples to estimate the population means. Supplementary materials accompanying this paper appear online.
- 23 August 2022
- Published by Springer Nature in Journal of Agricultural, Biological and Environmental Statistics
- Vol. 27 (4) , 652-673
Open AccessRow–Column Sampling Design Using Auxiliary Ranking Variables
In this paper, we present a new class of row–column sampling design when a sampling region is presented by non-overlapping quadrats of an rp-by-p row–column grid. Quadrats are rank-ordered on a relevant auxiliary variable, first in the rp rows of the grid and then in the so formed p ranked columns. This ranking enhances the precision of the sampling using ideas from standard rank set sampling methods. The sampling design falls in the traditional sampling framework where the sample selection probabilities are independent of the variables of interest. Selection probabilities are governed by a spatial design and the auxiliary information to facilitate better sampling coverage over the sampling region. The paper constructs unbiased estimators for population mean, total and provides variance estimates for them. It also shows when the new sampling design performs better than its competitors. The new sampling design is applied to a forest and two agricultural field samples to estimate the population means. Supplementary materials accompanying this paper appear online.
- 2 August 2022
- Published by Springer Nature
- p. 21-43
Judgment Post-stratified Sampling with Multiple Ranking: A Comparison with Ranked Set Sampling
- 2 August 2022
- Published by Springer Nature
- p. 21-43
Judgment Post-stratified Sampling with Multiple Ranking: A Comparison with Ranked Set Sampling
- 13 September 2021
- Published by Wiley in Canadian Journal of Statistics
- Vol. 50 (3) , 911-927
Combining ranking information from different sources in ranked‐set samples
This article considers a ranked‐set sample (RSS) with multiple ranking methods. The RSS data have K sets of ranks, each of which leads to a different estimator. We provide a weighted estimator that combines ranking information from these K sets of ranks. The weights are computed from the agreement scores of the ranking methods. We show that the new estimator provides substantial improvement over a balanced RSS estimator. The new estimator is applied to seed emergence data from a field experiment in agricultural research.
- 13 September 2021
- Published by Wiley in Canadian Journal of Statistics
- Vol. 50 (3) , 911-927
Combining ranking information from different sources in ranked‐set samples
This article considers a ranked‐set sample (RSS) with multiple ranking methods. The RSS data have K sets of ranks, each of which leads to a different estimator. We provide a weighted estimator that combines ranking information from these K sets of ranks. The weights are computed from the agreement scores of the ranking methods. We show that the new estimator provides substantial improvement over a balanced RSS estimator. The new estimator is applied to seed emergence data from a field experiment in agricultural research.