Estimation of ordinal population with multi-observer ranked set samples using ties information

2021 ◽  
pp. 096228022110259
Author(s):  
Amirhossein Alvandi ◽  
Armin Hatefi
Keyword(s):  
Likelihood Estimation ◽  
Simple Random Sampling ◽  
Ranked Set Sampling ◽  
Training Data ◽  
Estimation Of Parameters ◽  
Multiple Sources ◽  
Study Variable ◽  
Regression Approach ◽  
Ranking Information ◽  
Representative Samples

In many surveys, we often deal with situations where measuring the study variable is expensive; however, there are easy-to-measure characteristics which can be used as ranking information to obtain more representative samples from the population. Ranked set sampling is successfully employed in these cases as an alternative to commonly used simple random sampling. When the data is ordinal categorical, it is common to apply the ordinal logistic regression approach to ranked set sampling data for the estimation of parameters. This technique first depends on the information of training data. Besides, one is not capable of using the ranking information in the estimation process. In this paper, we propose a ranked set sampling scheme in which ranking information from multiple sources can be combined and incorporated efficiently into both data collection and estimation. The ranked set sampling data is used for non-parametric and maximum likelihood estimation of ordinal categorical population. Through extensive simulation studies, the performance of estimators is evaluated. The methods are finally applied to analyze bone disorder data and obesity data.

RATIO ESTIMATION OF THE POPULATION MEAN USING AUXILIARY INFORMATION UNDER THE OPTIMAL SAMPLING DESIGN

2020 ◽  
pp. 1-12
Author(s):  
Chunxian Long ◽  
Wangxue Chen ◽  
Rui Yang ◽  
Dongsen Yao
Keyword(s):  
Sampling Design ◽  
Auxiliary Information ◽  
Cost Effective ◽  
Simple Random Sampling ◽  
Auxiliary Variable ◽  
Ranked Set Sampling ◽  
Study Variable ◽  
Population Mean ◽  
Optimal Sampling Design ◽  
Extreme Ranked Set Sampling

Cost-effective sampling design is a problem of major concern in some experiments especially when the measurement of the characteristic of interest is costly or painful or time-consuming. In this article, we investigate ratio-type estimators of the population mean of the study variable, involving either the first or the third quartile of the auxiliary variable, using ranked set sampling (RSS) and extreme ranked set sampling (ERSS) schemes. The properties of the estimators are obtained. The estimators in RSS and ERSS are compared to their counterparts in simple random sampling (SRS) for normal data. The numerical results show that the estimators in RSS and ERSS are significantly more efficient than their counterparts in SRS.

scholarly journals Maximum likelihood estimation in location-scale families using varied L ranked set sampling

2020 ◽  
Author(s):  
Amer Al-Omari
Keyword(s):  
Maximum Likelihood ◽  
Random Sampling ◽  
Maximum Likelihood Estimators ◽  
Likelihood Estimation ◽  
Simple Random Sampling ◽  
Ranked Set Sampling ◽  
Sampling Schemes ◽  
Scale Parameters ◽  
Insight Into ◽  
Family Of Distributions

Recently, a generalized ranked set sampling (RSS) scheme has been introduced which encompasses several existing RSS schemes, namely varied L RSS (VLRSS), and it provides more precise estimators of the population mean than the estimators with the traditional simple random sampling (SRS) and RSS schemes. In this paper, we extend the work and consider the maximum likelihood estimators (MLEs) of the location and scale parameters when sampling from a location-scale family of distributions. In order to give more insight into the performance of VLRSS with respect to SRS and RSS schemes, the asymptotic relative precisions of the MLEs using VLRSS relative to that using SRS and RSS are compared for some usual location-scale distributions. It turns out that the MLEs with VLRSS are more precise than those with the existing sampling schemes.

scholarly journals Informative Bayesian model selection for RR Lyrae star classifiers

2021 ◽  
Vol 503 (1) ◽  
pp. 484-497
Author(s):  
F Pérez-Galarce ◽  
K Pichara ◽  
P Huijse ◽  
M Catelan ◽  
D Mery
Keyword(s):  
Machine Learning ◽  
Marginal Likelihood ◽  
Likelihood Estimation ◽  
Variable Stars ◽  
Training Data ◽  
Multiple Sources ◽  
Rr Lyrae Stars ◽  
Rr Lyrae ◽  
Lyrae Stars

ABSTRACT Machine learning has achieved an important role in the automatic classification of variable stars, and several classifiers have been proposed over the last decade. These classifiers have achieved impressive performance in several astronomical catalogues. However, some scientific articles have also shown that the training data therein contain multiple sources of bias. Hence, the performance of those classifiers on objects not belonging to the training data is uncertain, potentially resulting in the selection of incorrect models. Besides, it gives rise to the deployment of misleading classifiers. An example of the latter is the creation of open-source labelled catalogues with biased predictions. In this paper, we develop a method based on an informative marginal likelihood to evaluate variable star classifiers. We collect deterministic rules that are based on physical descriptors of RR Lyrae stars, and then, to mitigate the biases, we introduce those rules into the marginal likelihood estimation. We perform experiments with a set of Bayesian logistic regressions, which are trained to classify RR Lyraes, and we found that our method outperforms traditional non-informative cross-validation strategies, even when penalized models are assessed. Our methodology provides a more rigorous alternative to assess machine learning models using astronomical knowledge. From this approach, applications to other classes of variable stars and algorithmic improvements can be developed.

scholarly journals Even Order Ranked Set Sampling with Auxiliary Variable

2020 ◽  
Vol 18 (2) ◽  
Author(s):  
Muhammad Tayyab ◽  
Muhammad Noor ul-Amin ◽  
Muhammad Hanif
Keyword(s):  
Simple Random Sampling ◽  
Auxiliary Variable ◽  
Ranked Set Sampling ◽  
Sampling Scheme ◽  
Ratio Estimator ◽  
Study Variable ◽  
Median Ranked Set Sampling ◽  
Even Order ◽  
Efficient Alternative ◽  
The One

Even order ranked set sampling (EORSS) is a novel proposed ranked set sampling scheme connected with an auxiliary variable correlated with the study variable. This scheme quantifies only the one sampling unit which is at even position from each ranking set by employing specific criteria. The performance of the ratio estimator under EORSS is compared to its contemporary estimators in simple random sampling (SRS), ranked set sampling (RSS), median ranked set sampling (MRSS) and quartile ranked set sampling (QRSS) exploiting the same number of quantified units. The simulation results proved that EORSS is an efficient alternative sampling scheme for ratio estimation than SRS, RSS, MRSS and QRSS.

scholarly journals An Estimation of Parameters For Exponentiated Burr Type XII Distribution Based on Ranked Set Sampling

2020 ◽  
Vol 19 ◽  
pp. 26-34
Author(s):  
Abdallah Abdelfattah ◽  
Nagwan Reyad Elshendidy
Keyword(s):  
Random Sampling ◽  
Fisher Information Matrix ◽  
Information Matrix ◽  
Simple Random Sampling ◽  
Ranked Set Sampling ◽  
Estimation Of Parameters ◽  
Unknown Parameters ◽  
Method Of Maximum Likelihood ◽  
Mean Square Errors ◽  
Burr Type Xii

The aim of this paper is to estimate the parameters of exponentiated Burr type XII distribution (EBXII) based on ranked set sampling (RSS) technique, and also simple random sampling(SRS) is provided by the method of maximum likelihood. Fisher information matrix for both (SRS) and (RSS) for the unknown parameters are derived. Simulation study compared between the estimators of both methods in terms of their biases, mean square errors, and efficiencies. It is shown that the estimators based on RSS are more efficient than those of SRS.

scholarly journals Multi-Source Distilling Domain Adaptation

2020 ◽  
Vol 34 (07) ◽  
pp. 12975-12983
Author(s):  
Sicheng Zhao ◽  
Guangzhi Wang ◽  
Shanghang Zhang ◽  
Yang Gu ◽  
Yaxian Li ◽  
...  
Keyword(s):  
Domain Adaptation ◽  
Feature Space ◽  
Wasserstein Distance ◽  
Training Data ◽  
Source Distribution ◽  
Single Source ◽  
Multiple Sources ◽  
Target Domain ◽  
Target Feature ◽  
Training Samples

Deep neural networks suffer from performance decay when there is domain shift between the labeled source domain and unlabeled target domain, which motivates the research on domain adaptation (DA). Conventional DA methods usually assume that the labeled data is sampled from a single source distribution. However, in practice, labeled data may be collected from multiple sources, while naive application of the single-source DA algorithms may lead to suboptimal solutions. In this paper, we propose a novel multi-source distilling domain adaptation (MDDA) network, which not only considers the different distances among multiple sources and the target, but also investigates the different similarities of the source samples to the target ones. Specifically, the proposed MDDA includes four stages: (1) pre-train the source classifiers separately using the training data from each source; (2) adversarially map the target into the feature space of each source respectively by minimizing the empirical Wasserstein distance between source and target; (3) select the source training samples that are closer to the target to fine-tune the source classifiers; and (4) classify each encoded target feature by corresponding source classifier, and aggregate different predictions using respective domain weight, which corresponds to the discrepancy between each source and target. Extensive experiments are conducted on public DA benchmarks, and the results demonstrate that the proposed MDDA significantly outperforms the state-of-the-art approaches. Our source code is released at: https://github.com/daoyuan98/MDDA.

Stratified Ranked Set Sampling With Missing Observations for Estimating the Difference

2022 ◽  
pp. 209-232
Author(s):  
Carlos N. Bouza-Herrera
Keyword(s):  
Random Sampling ◽  
Real Data ◽  
Simple Random Sampling ◽  
Ranked Set Sampling ◽  
Missing Observations ◽  
Seminal Paper ◽  
Sample Error ◽  
The Difference ◽  
Stratified Ranked Set Sampling ◽  
Stratified Model

The authors develop the estimation of the difference of means of a pair of variables X and Y when we deal with missing observations. A seminal paper in this line is due to Bouza and Prabhu-Ajgaonkar when the sample and the subsamples are selected using simple random sampling. In this this chapter, the authors consider the use of ranked set-sampling for estimating the difference when we deal with a stratified population. The sample error is deduced. Numerical comparisons with the classic stratified model are developed using simulated and real data.

Sign in / Sign up

Export Citation Format

Share Document