Reliable Parameter Estimates Research Articles

THE IN-CONTROL PERFORMANCE OF THE ROBUST MULTIVARIATE SYNTHETIC CONTROL CHARTS FOR MONITORING MEAN SHIFT

Multivariate synthetic control chart enables monitoring of multiple process variable simultaneously and hence, negate the inflation of false alarm rates as can be seen in an individual statistical chart. A robust version of the synthetic chart is necessary to mitigate the issue faces by the traditional multivariate synthetic chart, which is unable to produce reliable parameter estimates when Phase I data are contaminated. Therefore, this study proposed three new robust multivariate synthetic control charts (CMRCD, CWS and CWP) which were constructed via minimum regularized covariance determinant (MRCD) and winsorized modified one-step M-estimator (WMOM) for efficient process monitoring. The effectiveness of the proposed robust charts was evaluated in terms of false alarm rates by comparing their in-control performances to the traditional multivariate synthetic chart, Cmean, which is based on the sample mean. Via extensive simulation studies, the findings indicate that the proposed robust control charts outperform the traditional chart regardless of the dimensions or the level of contaminations in the dataset. The real data study further validates that the CMRCD, CWS and CWP perform better than the Cmean. Specifically, the three robust control charts show significant observations, illustrating better capability in monitoring river water quality when compared to the Cmean, i.e., the traditional multivariate synthetic chart.

Optimal Estimation of Reliability Parameters for Modified Frechet-Exponential Distribution Using Progressive Type-II Censored Samples with Mechanical and Medical Data

The aim of this research is to estimate the parameters of the modified Frechet-exponential (MFE) distribution using different methods when applied to progressive type-II censored samples. These methods include using the maximum likelihood technique and the Bayesian approach, which were used to determine the values of parameters in addition to calculating the reliability and failure functions at time t. The approximate confidence intervals (ACIs) and credible intervals (CRIs) are derived for these parameters. Two bootstrap techniques of parametric type are provided to compute the bootstrap confidence intervals. Both symmetric loss functions such as the squared error loss (SEL) and asymmetric loss functions such as the linear-exponential (LINEX) loss are used in the Bayesian method to obtain the estimates. The Markov Chain Monte Carlo (MCMC) technique is utilized in the Metropolis–Hasting sampler approach to obtain the unknown parameters using the Bayes approach. Two actual datasets are utilized to examine the various progressive schemes and different estimation methods considered in this paper. Additionally, a simulation study is performed to compare the schemes and estimation techniques.

Performance of the Classical Model in Feature Selection Across Varying Database Sizes of Healthcare Data

Machine learning is increasingly being applied to medical research, particularly in selecting predictive modelling variables. By identifying relevant variables, researchers can improve model accuracy and reliability, leading to better clinical decisions and reduced overfitting. Efficient utilization of resources and the validity of medical research findings depend on selecting the right variables. However, few studies compare the performance of classical and modern methods for selecting characteristics in health datasets, highlighting the need for a critical evaluation to choose the most suitable approach. We analysed the performance of six different variable selection methods, which includes stepwise, forward and backward selection using p-value and AIC, LASSO, and Elastic Net. Health-related surveillance data on behaviors, health status, and medical service usage were used across ten databases, with sizes ranging from 10% to 100%, maintaining consistent outcome proportions. Varying database sizes were utilized to assess their impact on prediction models, as they can significantly influence accuracy, overfitting, generalizability, statistical power, parameter estimation reliability, computational complexity, and variable selection. The stepwise and backward AIC model showed the highest accuracy with an Area under the ROC Curve (AUC) of 0.889. Despite its sparsity, the Lasso and Elastic Net model also performed well. The study also found that binary variables were considered more crucial by the Lasso and Elastic Net model. Importantly, the significance of variables remained consistent across different database sizes. The study shows that no major variations in results between the fitness metric of the model and the number of variables in stepwise and backward p-value models, irrespective of the database's size. LASSO and Elastic Net models surpassed other models throughout various database sizes, and with fewer variables.

Evolving techniques for enhanced estimation: A comprehensive survey of stratified sampling and post-stratification methods

This comprehensive review delves into the landscape of estimators used in stratified sampling and post-stratification to estimate crucial population parameters, such as mean, median, variance, proportion, and distribution function. This study extensively examines the utilization of single and double auxiliary information, calibration weights, and various classes of estimators in the context of survey sampling, shedding light on their strengths and limitations. While these methods contribute significantly to enhancing estimation accuracy within stratified populations, this Review underscores challenges such as the dependence on accurate auxiliary information for single and double auxiliary methods, the assumption of ignorable nonresponse in calibration weights, and the need for careful consideration of underlying assumptions in a class of estimators. Future work in this field should focus on refining and validating these methodologies, addressing potential biases, and exploring innovative ways to handle nonresponses. Rigorous sensitivity analyses and investigations into the robustness of these techniques under varying conditions will be essential for advancing the precision and reliability of parameter estimation in the complex landscape of stratified sampling and post-stratification. This Review serves as a foundational resource for researchers and practitioners engaged in advancing the field and navigating the nuances of precise parameter estimation in survey sampling.

Bayesian parameter inference for epithelial mechanics

Cell-based mechanical models, such as the Cell Vertex Model (CVM), have proven useful for studying the mechanical control of epithelial tissue dynamics. We recently developed a statistical method called image-based parameter inference for formulating CVM model functions and estimating their parameters from image data of epithelial tissues. In this study, we employed Bayesian statistics to improve the utility and flexibility of image-based parameter inference. Tests on synthetic data confirmed that both our non-hierarchical and hierarchical Bayesian models provide accurate estimates of model parameters. By applying this method to Drosophila wings, we demonstrated that the reliability of parameter estimation is closely linked to the mechanical anisotropies present in the tissue. Moreover, we revealed that the cortical elasticity term is dispensable for explaining force-shape correlations in vivo. We anticipate that the flexibility of the Bayesian statistical framework will facilitate the integration of various types of information, thereby contributing to the quantitative dissection of the mechanical control of tissue dynamics.

Meta-analysis of genetic parameters for economic traits in rabbit using a random-effects model

The genetic improvement of rabbits helps increase their productivity and, consequently, increase the supply of animal protein for human consumption. The aim of this study was to perform a meta-analysis of genetic parameters (heritability and genetic correlation) for litter size at birth, litter weight at birth, litter size at weaning, litter weight at weaning and slaughter weight in rabbits. The final dataset contained 147 estimates of heritability and 32 estimates of genetic correlation across 34 articles published between 1992 and 2022. A random-effects model was used and the heterogeneity of estimates was assessed using Q and I2 statistics. Heritability estimates were of low magnitude for all traits, ranging from 0.09 to 0.18. The lowest heritability estimate was observed for litter size at weaning and the highest for slaughter weight. Most genetic correlations between traits were positive and moderate, ranging from 0.44 to 0.60. Significant heterogeneity among studies justified the use of random-effects models. The meta-analysis study provided reliable genetic parameter estimates and these results can support the development of rabbit breeding programmes..

Extension of the characterization of non-Gaussianity in gravitational wave detectors with a statistical hypothesis test

Abstract In gravitational wave (GW) astronomy, non-Gaussian noise, such as scattered light noise disturbs stable interferometer operation, limiting the interferometer’s sensitivity, and reducing the reliability of the analyses. In scattered light noise, the non-Gaussian noise dominates the sensitivity in a low frequency range of less than a few hundred Hz, which is sensitive to GWs from compact binary coalescence. This non-Gaussian noise prevents reliable parameter estimation, since several analysis methods are optimized only for Gaussian noise. Therefore, identifying data contaminated by non-Gaussian noise is important. In this work, we extended the conventional method to evaluate non-Gaussian noise, the Rayleigh statistic, by using a statistical hypothesis test to determine a threshold for non-Gaussian noise. First, we estimated the distribution of the Rayleigh statistic against Gaussian noise, called the background distribution, and validated that our extension serves as the hypothetical test. The threshold on the Rayleigh statistic is estimated at 0.73 and 1.28 when the significance level is 0.05, and the sample size is 39. Moreover, we investigated the detection efficiency by assuming two non-Gaussian noise models. For example, for the model with strong scattered light noise, the true positive rate (TPR) was always above 0.7 when the significance level was 0.05. For the demonstration, we applied our extension with the estimated thresholds to the real data. We confirmed our extension provides the binarized outputs of the statistical tests and contributes to finding the non-Gaussian noise in the real data. The results showed that our extension can contribute to an initial detection of non-Gaussian noise and lead to further investigation of the origin of the non-Gaussian noise.

Beyond Risk: A Measure of Distribution Uncertainty

This paper addresses the critical yet often overlooked concept of distribution uncertainty (ambiguity) in decision making, emphasizing its importance alongside traditional outcome uncertainty (risk). It introduces a novel quantitative measure of ambiguity that accurately captures distribution uncertainty. This measure enhances empirical models, yielding more reliable parameter estimates and improving decision-making processes. The study demonstrates the practical value of this ambiguity measure using financial market decision making as an example. The measure helps identify and adjust for uncertainties in underlying distributions, supporting more robust financial models and better risk management. The findings advocate for integrating ambiguity considerations into data analytics models and developing more reliable methodologies for empirical research and practical applications. This study promotes a nuanced understanding of uncertainty, offering significant implications for research methodologies and practical risk management across various fields.

A non-parametric model of ground motion parameters for shallow crustal earthquakes in Europe

The current study focuses on deriving ground motion models (GMMs) for 21 ground motion parameters derived from data sourced from the Engineering Strong Motion (ESM) database. These parameters include Peak Ground Acceleration (PGA), Peak Ground Velocity (PGV), Peak Ground Displacement (PGD), PGV-to-PGA ratio, (V/H) PGA ratio Predominant Frequency (Fp), Central Frequency (Ω), Spectral Parameter (q), Significant Duration (TSig), Root Mean Square Acceleration (Arms), Arias Intensity (Ia), Cumulative Absolute Velocity (CAV), Characteristic Intensity (IC), Acceleration Spectrum Intensity (ASI), Velocity Spectrum Intensity (VSI), Total Energy (Eacc), Spectral Centroid (Ew), Spectral Standard Deviation (Sw), Temporal Centroid (Et), Temporal Standard Deviation (St), and Correlation between time and frequency [ρ(t,ω)]. Both horizontal and vertical components are considered in this study. The inherent random effects within ground motion regression, encompassing inter-event, inter-site, inter-locality, and inter-region variabilities, are addressed using cross-nested mixed effect regression utilizing a non-parametric GMM approach employing Artificial Neural Network (ANN). Quantitative assessment of the models involves correlation coefficients for regression through the origin and error measures like mean squared error and mean absolute error. These findings of the assessment confirm reliable estimates of Ground Motion Parameters (GMPs). A comparison of GMPs computed using the proposed model and those reported in the literature indicated model's superior performance. Furthermore, satisfactory performance of the proposed GMM in ground motion simulation for the ESM region is demonstrated.

Distributional modelling of positively skewed data via the flexible Weibull extension distribution

SummaryThe time until an event occurs is often known to have a skewed distribution. To model this, a statistical distribution called the two‐parameter flexible Weibull extension (FWE) has been proposed. In this paper, the FWE distribution is used to model datasets through the use of generalised additive models for location, scale and shape (GAMLSS) distributional regression. GAMLSS is the only regression technique that can examine the effects of both categorical and numeric predictors on all the parameters of the distribution used to fit the dependent variable. To make it easier to use the FWE distribution through GAMLSS, the RelDists R package is proposed. A simulation study shows that FWE modelling through GAMLSS provides reliable parameter estimates even in the presence of factors that affect the distribution.

Diagnosis of Polymer Electrolyte Fuel Cell: Should We Rely Only on Electrical Quantities?

Polymer electrolyte fuel cells (PEMFC) will play a central role in the establishment of a de-carbonized society. They are seen as a key technology to empower electric heavy-duty vehicles. However, the lifespan is still below the fixed standards for their commercialization. A way to increase their durability is to complement them during operation with a mitigation strategy enabling to avoid the arise of faulty states accelerating degradation processes. Nowadays, most of the proposed fault identification and isolation architectures rely on monitoring of the performance through electrochemical impedance spectroscopy (EIS). Nevertheless, the analysis of impedance data is often complicated, since the contribution of many processes in the EIS spectra are coupled with each other and, for this reason, difficult to identify. This constitutes a limiting factor for its employment as diagnostic tool. In the last years, in order to overcome such issues, the use of nonelectrical periodic inputs and/or outputs has been advanced. The idea motivating this research direction is the possibility to detect the contribution of selected processes on the performance of PEMFCs separating them from the others. Novel frequency response techniques involving back pressure, temperature and partial pressure have been developed [1].In this contribution, the so-called concentration frequency response analysis (CFRA) is considered [2, 3]. It consists on perturbing the cathode side of the cell with a feed characterized by a periodic partial pressure of oxygen and water, and the detection of the electric response, voltage or current. A test station has been designed and constructed in order to apply this technique and investigate its capabilities. A model based analysis of the measured CFRA spectra indicates the possibility to decouple the contributions of the water transport into Nafion and oxygen mass transport on the performance of the cell depending on the chosen concentration input and electrical output. In this way, the analysis and identification of the related power losses is simplified with respect to classic EIS.Moreover, the performance of a fault identification and isolation architecture employing CFRA as monitoring technique have been evaluated and compared to EIS. For such purpose, an identifiability analysis of parameters related to faulty states and obtained by model fitting to spectra of each technique has been performed. The results clearly show that the employment of CFRA based on water pressure input offers the most reliable parameter estimation and, consequently, constitutes the most preferable diagnosis method among the others analyzed.

Optimizing pseudo-spiral sampling for abdominal DCE MRI using a digital anthropomorphic phantom.

For reliable DCE MRI parameter estimation, k-space undersampling is essential to meet resolution, coverage, and signal-to-noise requirements. Pseudo-spiral (PS) sampling achieves this by sampling k-space on a Cartesian grid following a spiral trajectory. The goal was to optimize PS k-space sampling patterns for abdomin al DCE MRI. The optimal PS k-space sampling pattern was determined using an anthropomorphic digital phantom. Contrast agent inflow was simulated in the liver, spleen, pancreas, and pancreatic ductal adenocarcinoma (PDAC). A total of 704 variable sampling and reconstruction approaches were created using three algorithms using different parametrizations to control sampling density, halfscan and compressed sensing regularization. The sampling patterns were evaluated based on image quality scores and the accuracy and precision of the DCE pharmacokinetic parameters. The best and worst strategies were assessed in vivo in five healthy volunteers without contrast agent administration. The best strategy was tested in a DCE scan of a PDAC patient. The best PS reconstruction was found to be PS-diffuse based, with quadratic distribution of readouts on a spiral, without random shuffling, halfscan factor of 0.8, and total variation regularization of 0.05 in the spatial and temporal domains. The best scoring strategy showed sharper images with less prominent artifacts in healthy volunteers compared to the worst strategy. Our suggested DCE sampling strategy also showed high quality DCE images in the PDAC patient. Using an anthropomorphic digital phantom, we identified an optimal PS sampling strategy for abdominal DCE MRI, and demonstrated feasibility in a PDAC patient.

The Jossens isotherm: A critical look at water contaminant adsorption modeling

ABSTRACT The Jossens isotherm is a useful model for describing water contaminant adsorption. This article critically examines several issues that impede its practical application. Initially, it clarifies the common misidentification of the Redlich–Peterson isotherm as the Jossens isotherm. Next, it addresses the challenge of deriving unique parameter values from the original loading-implicit form of the Jossens isotherm due to the high correlation between parameters. To resolve this problem, the article proposes a more robust solution: a loading-explicit form based on the Lambert W function, which facilitates reliable parameter estimation. Finally, the article critiques the current linearized Jossens isotherm, which relies on a trial-and-error method for parameter estimation, and introduces two alternative linearized forms that eliminate the need for this subjective approach.

Geometrical characterization of healthy red blood cells using digital holographic microscopy and parametric shape models for biophysical studies and diagnostic applications

Modeling of the red blood cell (RBC) shape is an integral part of the experimental and computer simulation investigations of light scattering by these cells for fundamental studies as well as diagnostic applications in techniques like cytometry and quantitative phase imaging. In the present work, a comprehensive study of the geometrical characterization of healthy human RBCs using digital holographic microscopy (DHM) and six frequently employed parametric shape models is reported. It is shown that the comparison of the optical phase profiles, and the thickness profiles given by the models with the DHM results gives a better judgment of the appropriateness of the parametric shape models. It is also shown that the RBC parametric models offer a simpler solution to the refractive index-thickness decoupling problem in QPI methods. Results of geometrical characterization of 500 healthy RBCs in terms of volume, surface area (SA), and sphericity index (SI) led to the classification of the parametric models in two categories based on the nature of variation of these quantities with the cell diameter. In light of the variability of the healthy RBC shapes, our findings suggest that the parametric models exhibiting a negative correlation between the SI and the cell diameter would provide more reliable estimates of the RBC parameters in diagnostic applications. Statistical distributions and descriptive statistics of the RBC volume, SA and SI serve as a guide for the assessment of the capability of the studied parametric models to give a reliable account of the variability of the healthy RBC shape and size.

Development of Mass-Conserving Atomistic Mathematical Model for Batch Anaerobic Digestion: Framework and Limitations

A variety of mathematical models have been developed to simulate the biochemical and physico-chemical aspects of the anaerobic digestion (AD) process to treat organic wastes and generate biogas. However, all these models, including the most widely accepted and implemented Anaerobic Digestion Model No.1, remain incapable of adequately representing the material balance of AD and are therefore inherently incapable of material conservation. The absence of robust mass conservation constrains reliable estimates of any kinetic parameters being estimated by regression of empirical data. To address this issue, the present work involved the development of a “framework” for a mass-conserving atomistic mathematical model which is capable of mass conservation, with a relative error in the range of machine precision value and an atom balance with a relative error of ±0.02% whilst obeying the Henry’s law and electroneutrality principle. Implementing the model in an Excel spreadsheet, the study calibrated the model using the empirical data derived from batch studies. Although the model shows high fidelity as assessed via inspection, considering several constraints including the drawbacks of the model and implementation platform, the study also provides a non-exhaustive list of limitations and further scope for development.

Analysis of Groundwater Time Series With Limited Pumping Information in Unconfined Aquifer: Response Function Based on Lagging Theory

AbstractGroundwater extraction from aquifers is a common practice for human use, and variations in groundwater levels can provide valuable information on the hydrogeological properties of the aquifer. However, reliable data on pumping rates and distribution are often lacking due to unsupervised groundwater pumping activities. This study presents a new mathematical model for transfer function modeling that depicts the drawdown response caused by pumping in an unconfined aquifer system. To account for the dense and unsupervised pumping events, the uniform pumping approach was used to estimate these effects. To more accurately represent unconfined flow, the model first integrates lagging theory into a response function derived from the Boussinesq equation. The lagging theory accounts for the effects of both inertial force and capillary suction. Furthermore, the model has been used to derive both specific yield and transmissivity along with two lagging parameters simultaneously using only groundwater level information from the Choshui River region in Taiwan. The estimated results suggest that this approach provides reliable estimates of hydrogeological parameters, demonstrating its usefulness for water resource management and water budget evaluation.

Reliability Modeling and Parameter Estimation for High-Speed Train Wheels Subject to Multi-Dimensional Degradation Processes Considering Mutual Dependency

Reliability Modeling and Parameter Estimation for High-Speed Train Wheels Subject to Multi-Dimensional Degradation Processes Considering Mutual Dependency

Construction of Twenty-Three Points Second Order Rotatable Design in Three Dimensions Using Trigonometric Functions

In this paper, a novel twenty-three-point second-order rotatable design is formulated utilizing trigonometric functions. Design of experiments plays a crucial role in various industries and research fields to investigate the relationship between multiple variables and their effects on a response variable. In particular, second order rotatable designs are widely used due to their ability to efficiently estimate the main, interaction, and curvature effects. Nevertheless, creating designs with a substantial number of points poses difficulties. This study concentrates on developing a second-order rotatable design with twenty-three points utilizing trigonometric functions. Trigonometric functions offer a systematic approach to distribute the points uniformly in the design space, thereby ensuring the optimal coverage of the experimental region. The proposed construction utilizes the properties of sine and cosine functions to generate a balanced and efficient design. The methodology involves dividing the design space into equidistant sectors and assigning the points using the trigonometric functions. By carefully selecting the starting angle and the angular increment, a complete and orthogonal design is achieved. The design is rotatable, meaning it can be rotated to any desired orientation without impairing the statistical properties of the design. Through this construction, the design effectively captures the main effects, interaction effects, and curvature effects. This enables reliable estimation of the model parameters, leading to accurate predictions and efficient optimization. Additionally, the design is efficient in terms of minimizing the number of experimental runs required, thereby reducing costs and time. The suggested second-order rotatable design comprising twenty-three points and employing trigonometric functions exhibits its superiority when compared to conventional designs. It offers a systematic and straightforward approach to construct a balanced and efficient design for studying the relationships between multiple variables. The design's rotatability ensures flexibility in experimental settings, making it a valuable tool for researchers and practitioners in various fields.

Dynamical study of Malaria epidemic: Stability and cost-effectiveness analysis in the context of Ghana

The Malaria epidemic is a serious public health issue that is worrying to the affected people in the tropics and subtropical parts of the world. The alarming consequences of the disease have prompted the scientific communities to seek better ways to manage the disease. In Ghana, the accurate statistical estimate of the Malaria transmission parameters is paramount to the public Health to embark on a transmission reduction program but rarely exists. The study’s purpose is to undertake a comprehensive mathematical analysis of a newly designed compartmentalized vector-host Malaria model capturing the aquatic mosquito vector compartment and vaccinated human host for the goal of estimating the model’s parameters and the threshold R0 computation and examining the associated bifurcation type that invokes at the Malaria-free state. The study thoroughly analyses the asymptomatic stability of the model to classify the respective behaviours at the equilibria. A sensitivity study of LHS-PRCC was undertaken to measure the variability in the threshold R0. Further, the model was subjected to a comprehensive analysis to assess the impact of vaccines on the dynamics of the disease. The analysis showed that vaccines have a substantial impact in reducing the Malaria cases in Ghana. To have a more reliable and accurate estimation of the model parameters of Ghana and robust mitigating intervention protocols for public health officials, the Least square fitting approach is applied to data of Ghana for an extended period of 10 years, starting from 2010 to 2020, to estimate the model’s parameters. The Malaria model was further structured into an intervention model to provide tailored eradication strategies for curtailing the disease. To exemplify the cost-effective procedure for assessing the cost related to the identified interventions, the ACER and ICER were used to analyse the cost of each intervention per the benefit. The results from the analyses suggest that increasing accessibility and executing a control intervention of 8 as identified by the cost analysis by Public Health will assist in mitigating the disease with a comparatively minimal cost.

A Hierarchical Bayesian approach to small area estimation of health insurance coverage in Ethiopian administrative zones for better policies and programs

Sample surveys are extensively used to provide reliable direct estimates for large areas or domains with enough sample sizes at national and regional levels. However, zones are unplanned domains by the Demographic and Health Survey (DHS) program and need more sample sizes to produce direct survey estimates with adequate precision. Conducting surveys in small areas (like zones) is too expensive and time-consuming, making it unfeasible for developing countries like Ethiopia. Therefore, this study aims to use the Hierarchical Bayes (HB) Small Area Estimation (SAE) model to estimate the Community-Based Health Insurance (CBHI) coverage at the zone levels in Ethiopia. To achieve this, we combined the 2019 Ethiopia Mini-Demographic and Health Survey (EMDHS) data with the 2007 population census data. SAE has addressed the challenge of producing reliable parameter estimates for small or even zero sample sizes across Ethiopian zones by utilizing auxiliary information from the population census. The results show that model-based estimates generated by the SAE approach are more accurate than direct survey estimates of CBHI. A map of CBHI scheme coverage was also used to visualize the spatial variation in the distribution of CBHI scheme coverage. From the CBHI scheme coverage map, we noticed notable variations in CBHI scheme coverage across Ethiopian zones. Additionally, this research identified areas with high and low CBHI scheme coverage to improve decision-making and increase coverage in Ethiopia. One of the novelties of this paper is estimating the non-sampled zones; therefore, the policymakers will give equal attention similar to the sampled zones.