Correlated Measurement Errors Research Articles

On some robust imputation methods in presence of correlated measurement errors with real data applications

Missing data occurs frequently in sample surveys. Numerous imputation techniques have been proposed to tackle the problem of missing data, and, in fact, limited works are available in the literature to deal with the issue of missingness while the data are ridden with measurement errors (ME). In addition, no work has been done on the robustness to handle the issue of missing data when it is contaminated with correlated measurement errors (CME). This article proposes some robust imputation methods (RIM) to impute the missing data in the presence of CME. The mean square error (MSE) of the proposed RIM is derived from the first-order approximation and examined with the MSE of the conventional imputation methods. The results are theoretically established and explained using a vast simulation study. Two applications of real data sets are presented to illustrate the efficiency and superiority of the suggested estimators relative to some estimators considered in this study.

Measurement errors and implications for preprocessing in miniaturised near-infrared spectrometers: Classification of sweet and bitter almonds as a case of study

Near-infrared (NIR) spectroscopy is a well-established analytical technique that has been used in many applications over the years. Due to the advancements in the semiconductor industry, NIR instruments have evolved from benchtop instruments to miniaturised portable devices. The miniaturised NIR instruments have gained more interest in recent years because of the fast and robust measurements they provide with almost no sample pretreatments.However, due to the very different configurations and characteristics of these instruments, they need a dedicated optimization of the measurement conditions, which is crucial for obtaining reliable results. To comprehensively grasp the capabilities and potentials offered by these sensors, it is imperative to examine errors that can affect the raw data, which is a facet frequently overlooked. In this study, measurement error covariance and correlation matrices were calculated and then visually inspected to gain insight into the error structures associated with the devices, and to find the optimal preprocessing technique that may result in the improvement of the models built.This strategy was applied to the classification of sweet and bitter almonds, which were measured with the three portable low-cost NIR devices (SCiO, FlameNIR+ and NeoSpectra Micro Development Kit) after removing the shelled, since their classification is of utmost importance for the almond industry. The results showed that bitter almonds can be classified from sweet almonds using any of the instruments after selecting the optimal preprocessing, obtained through inspection of covariance and correlation matrices. Measurements obtained with FlameNIR + device provided the best classification models with an accuracy of 98 %. The chosen strategy provides new insight into the performance characterization of the fast-growing miniaturised NIR instruments.

Modified correlated measurement errors model for estimation of population mean utilizing auxiliary information

The existence of measurement errors cannot be avoided in practice. It is a prominent fact that the existence of measurement errors diminishes conventional properties of the estimators. A modified correlated measurement errors model has been proposed. Shalabh and Tsai (Commun Stat Simul Comput 46(7):5566–5593. 10.1080/03610918.2016.1165845, 2017) correlated measurement errors model is a particular member of the suggested modified model. In this article, we have tackled the estimation of population mean utilizing auxiliary information under modified correlated measurement errors model. We have developed ratio and product estimators and studied their properties in case of simple random sampling without replacement (SRSWOR) up to first order of approximation. It has been illustrated that suggested ratio and product estimators are more efficient than the conventional unbiased estimator as well as Shalabh and Tsai (Commun Stat Simul Comput 46(7):5566–5593. 10.1080/03610918.2016.1165845, 2017) ratio and product estimators under very realistic situations. An empirical study has also been performed to demonstrate the merits of the recommended estimators over other estimators.

Ranked set sampling imputation methods in presence of correlated measurement errors

Very little attention has been paid by the authors for providing improved imputation methods under correlated measurement errors (CMEs) using various sampling designs. This article addresses several power ratio imputation methods and the resulting estimators under ranked set sampling (RSS) when CMEs are present. The mean square error (MSE) is developed to assess how well the suggested estimators work under CMEs. The effectiveness of the proposed imputation methods and corresponding resultant estimators is assessed by a comprehensive simulation utilizing an artificially generated population. Further, an application of the proposed imputation methods is also provided using real data from Sweden’s municipalities consisting of the total number of seats in 1982 in municipal council taken as study variable and the number of conservative seats in 1982 in municipal council taken as auxiliary variable. The CMEs might arise due to common surveyor biases if the same surveyor collects information on both the “total number of seats in municipal council” and “the number of conservative seats in that municipal council”. The simulation and real data results show that the proposed power ratio imputation methods outperform the mean, conventional ratio, and product imputation methods.

Little evolution of dust emissivity in bright infrared galaxies from 2 < z < 6

ABSTRACT Variations in the dust emissivity index, $\beta$, within and between galaxies, are evidence that the chemistry and physics of dust must vary on large scales, although the nature of the physical and/or chemical variations is still unknown. In this paper, we estimate values of $\beta$ and dust temperature for a sample of 109 dusty star-forming galaxies (DSFGs) over the range, $2 \ \lt\ z \ \lt\ 6$. We compare the results obtained with both an optically thin model and a general opacity model, finding that our estimates of $\beta$ are similar between the models but our estimates of dust temperature are not. We find no evidence of a change in $\beta$ with redshift, with a median value of $\beta = 1.96$ for the optically thin model with a confidence interval (16–84 per cent) of 1.67 to 2.35 for the population. Using simulations, we estimate the measurement errors from our procedure and show that the variation of $\beta$ in the population results from intrinsic variations in the properties of the dust in DSFGs. At a fixed far-infrared luminosity, we find no evidence for a change in dust temperature, $T_{\textrm {dust}}$, with redshift. After allowing for the effects of correlated measurement errors, we find an inverse correlation between $\beta$ and $T_{\textrm {dust}}$ in DSFGs, for which there is also evidence in low-redshift galaxies.

Study of efficiency of stratified estimator of population mean under correlated measurement error

Study of efficiency of stratified estimator of population mean under correlated measurement error

Impact of Correlated Measurement Errors on Some Efficient Classes of Estimators

It is well-known that the appearance of measurement errors spoils the traditional results in survey sampling. The concept of correlated measurement errors (CMEs) is true in various practical situations, but this has been seldom considered by researchers in survey sampling. In this article, the influence of the CME under simple random sampling (SRS) has been considered over some prominent classes of estimators for the population mean. The first-order approximated formulae of the mean square error of the introduced estimators are reported, and a comparative analysis has also been conducted with traditional estimators. The theoretical findings are extended by a broad spectrum computational study using real and artificial data.

Multivariate longitudinal analysis for the association between brain atrophy and cognitive impairment in prodromal Huntington's disease subjects.

Cognitive impairment has been widely accepted as a disease progression measure prior to the onset of Huntington's disease. We propose a sophisticated measurement error correction method that can handle potentially correlated measurement errors in longitudinally collected exposures and multiple outcomes. The asymptotic theory for the proposed method is developed. A simulation study is conducted to demonstrate the satisfactory performance of the proposed two-stage fitting method and shows that the independent working correlation structure outperforms other alternatives. We conduct a comprehensive longitudinal analysis to assess how brain striatal atrophy affects impairment in various cognitive domains for Huntington's disease.

Impact assessment of correlated measurement errors using logarithmic-type estimators

In survey sampling, several estimation procedures have been proffered by various prominent authors to compute the impact of measurement errors (ME) but the impact of correlated measurement errors (CME) has been computed only by Shalabh and Tsai [Ratio and product methods of estimation of population mean in the presence of correlated measurement errors. Commun Stat Simul Comput. 2016;46(7):5566–5593]. This study provides a novel approach to compute the impact of CME through some logarithmic-type estimators using simple random sampling (SRS). The properties of the proffered estimators have been studied and compared with the properties of the conventional estimators. A numerical study and a broad spectrum simulation study are accomplished over real and artificially generated populations to support the theoretical results.

On classes of robust estimators in presence of correlated measurement errors

It is an established fact that the existence of measurement errors (ME) vitiates the conventional properties of the estimators. The impact of correlated measurement errors (CME) was first assessed by Shalabh and Tsai (2017) using classical ratio and product estimators. This article makes a modest attempt to suggest some classes of robust estimators of population mean in the presence of CME. The bias and mean square error of the suggested estimators are reported with the first-order approximation and the efficiency conditions are derived. The theoretical results are extended with the help of a simulation study executed over a hypothetically rendered population. Apart from this, the effect of ME on the suggested estimators is also studied both theoretically and empirically.

The role of structural social capital in driving social-oriented sustainable agricultural entrepreneurship

Sustainable agriculture and structural social capital (SSC) have emerged as critical players in socioeconomic recovery and sustainable food supply chains. Therefore, implementing transformation toward sustainable agriculture is necessary, considering situational conditions are crucial prerequisites. However, studies have yet to thoroughly examine the role of this social capital (SC) type in influencing sustainable agricultural entrepreneurship (SAE) under the socially impacted supply chain perspective. We clarify these critical knowledge gaps with the variable of facilitating conditions (PFCs) investigated as a mediator. To do so, this research adopts (1) established theoretical lenses in the form of SC and social entrepreneurship (SE) theories and theory of planned behavior (TPB); (2) a mixed-methods approach in the form of in-depth interviews and the survey; and (3) structural equation modeling using data collected from 416 agricultural entrepreneurs in Vietnam. This structural approach helps rigorously evaluate causal relationships based on a reliable scale validation process due to incorporating measurement error and correlated measurement error into the model ignored by some other popular approaches (e.g., regression equations). This research contributes to (1) the reconciliation of scales of the existing constructs, (2) the development of a new scale (i.e., PFCs to SAE), and (3) the discovery of relationships among SSC, PFCs, TPB's components, and intention to SAE (i.e., ISAE). The result shows the critical role of SSC in driving ISAE, and thus, it significantly contributes to the existing SC, TPB, and SE literature. It is also helpful for socially driven agricultural entrepreneurs in developing countries amid ongoing challenges.

Performance evaluation of novel logarithmic estimators under correlated measurement errors

ABSTRACT In survey research, the issue of measurement errors (ME) has been sorted out by various authors, but the issue of correlated measurement errors (CME) has only been studied by Shalabh and Tsai (2017) till date. This article provides a modest acquaintance to evaluate the performance of few novel logarithmic estimators of population mean in the existence of CME under simple random sampling (SRS). The mean square error of the proposed estimators has been obtained. It has been exhibited theoretically under certain conditions that the proposed class of estimators dominates their conventional counterparts. Furthermore, the theoretical findings have been assessed with a Monte Carlo simulation using a hypothetically gendered population and proper suggestions have been forwarded to the survey professionals.

Traffic Risk Image in The Light of the Road Situation Assessment Questionnaire

Introduction: Drivers often lack reflection on their behaviour. For predictive purposes and educational interventions, it is useful to know the individual traffic risk picture of the driver under investigation.
 Objective: This paper aimed to show the psychometric properties of the questionnaire and to establish a hierarchy of traffic situations in terms of perceptions of associated danger.
 Material and Methods: A total of 548 people aged between 18 and 68 years participated in the study. A self-administered questionnaire method was used to measure situational (road) awareness and the ability to anticipate the consequences of engaging in unsafe road traffic actions. The tool's accuracy was verified with the 'Controversy' questionnaire, which is used to measure the severity of drivers' normative beliefs about breaking traffic laws.
 Results: Satisfactory psychometric values of the questionnaire were obtained. After taking into account the correlation of measurement errors, an acceptable fit was obtained: χ2 /df = 2.48; GFI = 0.95; AGFI = 0.92; CFI = 0.96; RMSEA = 0.06 (0.048-0.072). Mean values and standard deviations for 14 descriptions of unsafe traffic situations were shown.
 Conclusions: The satisfactory reliability and accuracy of the measurement allow the method to be used in research and individual diagnosis in the areas of transport psychology and education. Overtaking at pedestrian crossings, on a bend, uphill, and continuing to drive despite fatigue are the most dangerous traffic situations associated with the risk of traffic accidents.

Logistic regression with correlated measurement error and misclassification in covariates.

Many areas of research, such as nutritional epidemiology, may encounter measurement errors of continuous covariates and misclassification of categorical variables when modeling. It is well known that ignoring measurement errors or misclassification can lead to biased results. But most research has focused on solving these two problems separately. Addressing both measurement error and misclassification simultaneously in a single analysis is less actively studied. In this article, we propose a new correction method for a logistic regression to handle correlated error variables involved in multivariate continuous covariates and misclassification in a categorical variable simultaneously. It is not computationally intensive since a closed-form of the approximate likelihood function conditional on observed covariates is derived. The asymptotic normality of this proposed estimator is established under regularity conditions and its finite-sample performance is empirically examined by simulation studies. We apply this new estimation method to handle measurement error in some nutrients of interest and misclassification of a categorical variable named physical activity in the European Prospective Investigation into Cancer and Nutrition-InterAct Study data. Analyses show that fruit is negatively associated with type 2 diabetes for a group of women doing active physical activity, protein has positive association with type 2 diabetes for the group of less active physical activity, and actual physical activity has a greater effect on reducing the risk of type 2 diabetes than observed physical activity.

Noise-resistant quantum state compression readout

Qubit measurement is generally the most error-prone operation that degrades the performance of near-term quantum devices, and the exponential decay of readout fidelity severely impedes the development of large-scale quantum information processing. Given these disadvantages, we present a quantum state readout method, named compression readout, that naturally avoids large multi-qubit measurement errors by compressing the quantum state into a single qubit for measurement. Our method generally outperforms direct measurements in terms of accuracy, and the advantage grows with the system size. Moreover, because only one-qubit measurements are performed, our method requires solely a fine readout calibration on one qubit and is free of correlated measurement error, which drastically diminishes the demand for device calibration. These advantages suggest that our method can immediately boost the readout performance of near-term quantum devices and will greatly benefit the development of large-scale quantum computing.

Quantifying the redshift space distortion of the bispectrum III : detection prospects of the multipole moments

ABSTRACT The redshift space anisotropy of the bispectrum is generally quantified using multipole moments. The possibility of measuring these multipoles in any survey depends on the level of statistical fluctuations. We compute the statistical fluctuations in the measurement of bispectrum multipoles for a Euclid like galaxy survey based on second-order perturbation theory and present two quantities: the signal-to-noise ratio (SNR) which quantifies the detectability of a multipole and the rank correlation which quantifies the correlation in measurement errors between any two multipoles. Based on SNR values, we find that Euclid can potentially measure the bispectrum multipoles up to ℓ = 4 across various triangle shapes, formed by the three k vectors in Fourier space. In general, SNR is maximum for the linear triangles. SNR values also depend on the scales and redshifts of observation. While, ℓ ≤ 2 multipoles can be measured with SNR > 5 even at linear/quasi-linear ($k_1 \lesssim 0.1 \, {\rm Mpc}^{-1}$) scales, for ℓ > 2 multipoles, we require to go to small scales or need to increase bin sizes. These estimates are based on bins of extent Δln k1 = 0.1, Δμ = 0.05, and Δt = 0.05, where k1 is the length of the largest side, and (μ, t), respectively, quantify the size and shape of the triangles. For most multipole pairs, the errors are only weakly correlated across much of the triangle shapes barring a few in the vicinity of squeezed and stretched triangles. This makes it possible to combine the measurements of different multipoles to increase the effective SNR.

HEARING LOSS AND SUBJECTIVE COGNITIVE DECLINE IN THE ARIC-NEUROCOGNITIVE STUDY

Abstract Subjective cognitive decline (SCD) impacts quality-of-life, healthcare utilization and predicts dementia. If hearing loss is associated with SCD is unknown. We estimated the cross-sectional association between hearing loss and SCD in 2,536 participants (79.3±4.4 years, 60% female, 20% Black) in the Atherosclerosis Risk in Communities Neurocognitive Study. SCD was defined as self-reported persistent decline in memory (yes/no) among adults with normal cognition. Hearing was measured using pure tone better-ear thresholds (0.5-4 kHz), speech-in-noise, and self-report. Using Poisson models with robust standard errors, after adjustment for demographic and clinical covariates, self-reported moderate or greater trouble hearing (vs. excellent/good) was associated with a 30% increase in the prevalence of SCD (95%CI:1.12,1.51). Audiometric hearing and speech-in-noise performance were not associated. Use of self-report to approximate audiometric hearing warrants caution, particularly when the outcome is also self-reported; our findings suggest the association between self-reported hearing and SCD may be due to correlated measurement error.

Hybrid estimators of population mean in presence of non-response and correlated measurement errors

The present work deals with examining the effect of measurement errors on the design based statistical properties of the estimators of the population mean when the survey is encountered with some non-response. The technique of sub-sampling of non-respondents have been equipped with calibration techniques to handle the presence of non-response. Two kind of distance measures have been proposed and studied in detail to improve the calibration weights. The data on study as well as auxiliary variable has been assumed to be contaminated with correlated measurement errors. The specific properties of the estimators are derived under simple random sampling without replacement design. The proposed hybrid estimators are supported with numerical illustration incorporating simulation using specific population to show the applicability of proposed techniques practically. It is observed that the absolute relative bias of the estimators increase slightly with increase in error variance. Similarly, the precision of the estimators degrade with increase in error dispersion. Overall the proposed hybrid estimator with the chi-square type distance measure is observed to be remarkably efficient in comparison to the logarithmic type distance measure in presence of non-response and measurement errors.

Analysis of Structural Health Monitoring Data with Correlated Measurement Error by Bayesian System Identification: Theory and Application.

Measurement error is non-negligible and crucial in SHM data analysis. In many applications of SHM, measurement errors are statistically correlated in space and/or in time for data from sensor networks. Existing works solely consider spatial correlation for measurement error. When both spatial and temporal correlation are considered simultaneously, the existing works collapse, as they do not possess a suitable form describing spatially and temporally correlated measurement error. In order to tackle this burden, this paper generalizes the form of correlated measurement error from spatial correlation only or temporal correlation only to spatial-temporal correlation. A new form of spatial-temporal correlation and the corresponding likelihood function are proposed, and multiple candidate model classes for the measurement error are constructed, including no correlation, spatial correlation, temporal correlation, and the proposed spatial-temporal correlation. Bayesian system identification is conducted to achieve not only the posterior probability density function (PDF) for the model parameters, but also the posterior probability of each candidate model class for selecting the most suitable/plausible model class for the measurement error. Examples are presented with applications to model updating and modal frequency prediction under varying environmental conditions, ensuring the necessity of considering correlated measurement error and the capability of the proposed Bayesian system identification in the uncertainty quantification at the parameter and model levels.

Analyzing the behavior of general class of estimators of population mean in presence of correlated measurement errors

Measurement error, which occurs when some variables of interest cannot be observed exactly. It often leads to invalid conclusions and lamentable implications if ignored. Therefore, the aim of the present work is to investigate ways to handle them when the observations on both the study and auxiliary variables are spoiled with measurement errors. A general class of estimators for the estimation of population mean in presence of correlated measurement errors on study and auxiliary variables has been considered. The variations in the properties of the estimators under the influence of measurement errors has been discussed in detail. The considered class of estimators is compared with the recently proposed family of exponential-type estimators. Simulation studies are carried out to show the impact on the behavior of estimators due to variation in parameters of the measurement error model.