Statistical Limits Research Articles

What factors preventing the older adults in China from living longer: a machine learning study

BackgroundThe fact that most older people do not live long means that they do not have more time to pursue self-actualization and contribute value to society. Although there are many studies on the longevity of the elderly, the limitations of traditional statistics lack the good ability to study together the important influencing factors and build a simple and effective prediction model.MethodsBased on the the data of Chinese Longitudinal Healthy Longevity Survey (CLHLS), 2008–2018 cohort and 2014–2018 cohort were selected and 16 features were filtered and integrated. Five machine learning algorithms, Elastic-Net Regression (ENR), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN), and eXtreme Gradient Boosting (XGBoost), were used to develop models and assessed by internal validation with CLHLS 2008–2018 cohort and temporal validation with CLHLS 2014–2018 cohort. Besides, the best performing model was explained and according to the variable importance results, simpler models would be developed.ResultsThe results showed that the model developed by XGBoost algorithm had the best performance with AUC of 0.788 in internal validation and 0.806 in temporal validation. Instrumental activity of daily living (IADL), leisure activity, marital status, sex, activity of daily living (ADL), cognitive function, overall plant-based diet index (PDI) and psychological resilience, 8 features were more important in the model. Finally, with these 8 features simpler models were developed, it was found that the model performance did not decrease in both internal and temporal validation.ConclusionsThe study indicated that the importance of these 8 factors for predicting the death of elderly people in China and built a simple machine learning model with good predictive performance. It can inspire future key research directions to promote longevity of the elderly, as well as in practical life to make the elderly healthy longevity, or timely end-of-life care for the elderly, and can use predictive model to aid decision-making.

특수교육 중재전략 연구에서 베이지안 종단자료 분석모형의 유용성 탐색

Research on the effectiveness of intervention strategies for students with special needs is an important research topic in the field of special education. Until now, the majority of special education research on this topic has been based on the use of frequentist statistics and pre- and post-test designs. However, recent research in the natural and social sciences has criticised the limitations of frequentist statistics and emphasised the importance of reliable longitudinal data. Therefore, the purpose of this study is to introduce a Bayesian longitudinal data analysis model that can overcome the limitations of frequentist statistics in special education situations where longitudinal data are collected. According to the results of this study, when using Bayesian longitudinal data analysis model to test the effectiveness of intervention strategies, it is not only possible to reflect previous studies in parameter estimation, but also to continuously update the effect size of existing intervention strategies. In addition, it is possible to estimate effect sizes with good reliability even in research conditions with a small number of cases, such as in the field of special education. Finally, the limitations of Bayesian theory and future research topics are discussed.

Analysis of chemical exposures in racial populations in Canada: An investigation based on the Canadian health measures survey

Despite demonstrated disparities in environmental chemical exposures by racial identity, no Canadian study has systematically assessed the feasibility of using a nationally representative dataset to examine differences in chemical concentrations by race. We assessed the feasibility and constraints of analysing chemical exposures in racial populations, including visible minorities and populations of Indigenous identity, using biomonitoring data collected through the Canadian Health Measures Survey (CHMS). Our primary objectives were to assess the ability to 1) generate geometric means and percentiles of chemical concentrations for racial populations by age or sex, 2) statistically compare concentrations among racial populations, and 3) calculate time trends of concentrations by race. We conducted these analyses for several priority chemicals: lead, cadmium, benzene, bisphenol A (BPA), and di(2-ethylhexyl) phthalate (DEHP). Survey participants self-identified as one of the following: White, Black, East and Southeast Asian, South Asian, Middle Eastern, Latin American, First Nations, Metis, and Inuit. Analyses were conducted for individual and combined cycles of the CHMS. Using data from the latest CHMS cycle in which each chemical was measured, we observed that sample sizes were sufficient to report geometric mean concentrations for all races except Inuit. Due to privacy considerations associated with small sample sizes, the 5th and 95th percentile concentrations could not be consistently reported for all racial populations in this analysis. While we were able to statistically compare concentrations among racial populations, the analysis was constrained by the limited number of statistical degrees of freedom available in a single CHMS cycle. Both of these constraints were alleviated by combining multiple cycles of data. The analysis of time trends was less subject to privacy and statistical limitations; we were able to calculate time trends of chemical concentrations for all racial populations. Our findings provide an important baseline for follow-up investigations of descriptive and etiological analyses of environmental chemical exposures and race in the CHMS.

Experimental analysis of scour around an offshore wind gravity base foundation

Scour processes and corresponding scour protection measures have been extensively studied for vertical cylinders. Several formulations for estimating scour and determining the optimum size of rocks for scour protection are widely available in the literature. However, when studying other types of structures, the geometric variability of the structures, limits the application of available semiempirical formulations. Scour phenomena around complex structures have been investigated experimentally in reduced-scale wave-current basins. However, the number of tests focused on support typologies different from monopile structures is still limited, restricting the availability of semiempirical formulations focused on conceptual design of scour protection solutions or in the estimation of the free scour development.In this work, a better understanding of the scour processes around gravity-based foundations (GBFs) and the performance of different scour protection solutions is gained via experimental results obtained from a large-scale test programme (1:35 test scale), including combined wave and current tests. The experimental tests were divided into two main types: (1) free scour and (2) scour protection tests. The main conclusion was that the triggering of a horseshoe vortex is the main driver of scour processes together with the contraction of flow lines at the foundation contours. Since monopile semiempirical formulations are not suitable for scour prediction, a new semiempirical approach was obtained from the test results. Moreover, two distinct scour mitigation methods were analysed (rock protection and concrete mattresses). Based on the undisturbed mobility parameter, statistical and dynamic limits were defined considering the rock grading tested. Finally, for concrete mattresses, the design driver was identified, and the adaptability of these types of solutions to seabed changes was verified.

Some results on $I_2$-deferred statistically convergent double sequences in fuzzy normed spaces

The primary objective of this study is to introduce the concepts of $I_2$-deferred Cesàro summability and $I_2-$ deferred statistical convergence for double sequences in fuzzy normed spaces (FNS). Furthermore, the aim is to explore the connections between these concepts and subsequently establish several theorems pertaining to the notion of $I_2$-deferred statistical convergence in FNS for double sequences. We further define $I_2$-deferred statistical limit points and $I_2$-deferred statistical cluster points of a sequence within FNS and explore the relationships among these concepts.

Power prediction using high-resolution SCADA data with a farm-wide deep neural network approach

Accurate loss estimation methods with a high level of temporal granularity are necessary to enable the implementation of efficient and adaptable control strategies for wind farms. Predictive models for the power of wind turbines within a wind farm are investigated using high-resolution SCADA data and deep learning methodologies. Traditional physical models offer detailed insights but are computationally expensive. Statistical models face limitations in handling wind energy variability. In this study, deep learning models are explored to capture spatial and temporal dynamics, with four models developed: Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory network (LSTM), and a hybrid CNN-LSTM model. SCADA data from an offshore wind farm is preprocessed, anomalies removed, and annotated based on operational regions. The models are trained, validated, and tested, demonstrating improved accuracy over baseline methods. The hybrid model, incorporating spatial and temporal information, yields the highest predictive performance, showcasing the significance of these dimensions in wind power prediction.

Generalization of PINNs for elliptic interface problems

In this letter, we investigate the statistical limits of deep learning for learning solutions of elliptic interface problems from randomly generated data by employing Physics-informed Neural Networks (PINNs). We prove a lower bound and an upper bound for the generalization error of PINNs for solving elliptic interface equations with the Dirichlet boundary condition. In particular, the upper bound on the generalization error of PINNs is obtained by computing the fixed points of the local Rademacher complexity. We show that variations in volume across distinct regions exert influence on the requisite sample complexity. This unveils an underlying principle within the sampling strategy: the sample size of the data is contingent upon the Lebesgue measure of the domain, coupled with the smoothness and dimensionality characteristics of the interface problem. This letter sheds light on the network architecture design for solving interface problems using PINNs.

Hopes and Limitations of Reproducible Statistics and Machine Learning

Hopes and Limitations of Reproducible Statistics and Machine Learning

Forecasting Limits of Feasible Organization Growth by Using Key Predictors of Organizational Transformation

The article studies methodological approach, advanced by the author, to using key predictors of organizational transformation necessary to forecast limits of feasible organization growth within the frames of the present production structure. The principle distinguishing feature of this methodological approach is the use of indicators defined by the author as key predictors of organizational transformation. This set of indicators was identified by the author through results of earlier statistical research of transformation laws of organization production structure in different sectors of economy. With the help of such indicators it is possible not only to describe the predominant type of organization production structure but also, on the basis of their quantitative values, to identify limits of corresponding level of its development. Methodological approach advanced by the author is based on statistical methods of research and gives an opportunity to forecast limits of feasible organization growth based on the earlier developed method of their production structure. Within the frames of the given methodological approach the method of statistical forecasting limits of feasible organization growth was elaborated. The use of this methodology could allow organization management to forecast in advance the moments of coming organizational transformation, which is connected with changes in production structure type and to make necessary managerial decisions.

How political and social constituent traits affect the responsiveness of legislators: A Comparative Field Experiment

AbstractA growing body of literature investigates whether legislators show biases in their constituency communication contingent upon constituent traits. However, we know little about whether and how findings of unequal responsiveness generalize across countries (beyond the United States) and across different traits. We address both issues using a pre‐registered comparative field experiment conducted in Germany, the United Kingdom and the Netherlands, in which fictitious citizens (varied by ethnicity, social class and partisan affiliation) inquired about legislators’ policy priorities regarding the COVID‐19 pandemic. Our pooled analysis reveals that co‐partisanship and class both increase the responsiveness of legislators while we find no effect for ethnicity. The effect sizes we find are small, but comparable to earlier studies and also noteworthy in view of our hard test design. Our exploratory analyses further corroborate the lack of discrimination against ethnic minority constituents in showing no intersectionality effects, that is, interactions between ethnic‐minority and low‐class identities. This exploratory step also addresses the country specific differences that we find. We speculate about plausible underlying party system effects that we, however, cannot substantiate due to statistical limitations. This important issue requires further attention in future research.

A Review of Brain Tumor Segmentation Using MRIs from 2019 to 2023 (Statistical Information, Key Achievements, and Limitations)

A Review of Brain Tumor Segmentation Using MRIs from 2019 to 2023 (Statistical Information, Key Achievements, and Limitations)

Estimating the as-built thermal performance of dwellings using simulated on-board data: From ideal to limited monitoring

Advancement in non-intrusive monitoring of energy usage in households presents an ideal opportunity to get insight into the as-built performance of buildings and assess performance indicators such as the Heat Loss Coefficient (HLC) of the building fabric. The availability of in-use monitored data contributes to bridging the energy performance gap while avoiding costly and exhaustive dedicated measurement campaigns. Methodologies to assess the building performance by combining on-board monitoring and statistical data-driven models are widely accepted. However, their unsupervised and automated applicability on a large scale, with limited data and no insight into the building characteristics and occupants’ behaviour still needs to be investigated. Therefore, this work aims at assessing the reliability and performance of statistical black-box AutoRegressive with eXogenous input (ARX) models on artificial measurement campaigns starting from idealistic setups and progressing towards limited ones. In this investigation, the uncertainty in the estimate inherited from inputs, such as different heat sources or weather data, and their post-processing is combined with the uncertainty originating from different monitoring packages, which results in a thorough overview of the statistical modelling limitations. Results show that disaggregation of the final heating usage poses a major challenge. In cases where gross gas usage is measured, and the system efficiency and domestic hot water consumption are not taken into account, the HLC can deviate by 49% in low-performing houses and by 106% in well-performing houses. However, by applying data pre-processing methods the discrepancy can be lowered to 8% and 23% respectively, but with high uncertainty. On the other hand, the installation of an additional heat meter could yield deviations from the target of 26% for low-performing and 11% for well-performing dwellings without data manipulation, both with low uncertainties.

Assessment of Heart Rate Variability in Children with Mitral Valve Prolapse.

Mitral valve prolapse (MVP) is a heart valve anomaly with specific electrocardiographic findings and arrhythmia. A predominant sympathetic tone and diminished vagal activity have been reported especially in symptomatic MVP patients. In the current study, we aim to review heart rate variability (HRV) parameters of MVP children in order to determine if there is an impaired autonomic regulation. The data of children with MVP were retrospectively analyzed. Demographic characteristics, body mass index, symptomatology, MVP type and mitral regurgitation, MVP-related electrocardiographic changes, arrhythmia, and HRV parameters in 24-h Holter monitoring were recorded. HRV parameters of the control group were obtained from children applying for sport participation. Statistical significance limit was accepted as p < 0.05. 154 of the patients (74.8%) were girls, the median age was 13.58 ± 2.99years. While MVP patients had shorter PR intervals, those who had syncope had longer PR intervals. Short PR distance may be a feature of MVP, long PR distance may be a condition associated with syncope with the underlying paroxysmal atrioventricular block. Moreover, the minimum heart rate was significantly lower in the MVP group compared to the control and there were no significant differences in terms of remaining parameters. The number of low LF values was higher in MVP patients than the control. Comparing HRV values of groups by gender, we found that overall HRV parameters were lower in girls with MVP while minimum, maximum, and average heart rate were lower in boys with MVP. Impaired HRV associated with MVP could be age, gender, and symptom related. In addition, low LF may indicate impaired baroreflex sensitivity in MVP patients.

Control charting methods for monitoring high dimensional data streams: A conceptual classification scheme

There are always challenges in various industrial or non-industrial processes in which the product quality/service is described by a large number of quality characteristics. Thus, statistical process monitoring (SPM) techniques for capturing the quality of high-dimensional processes are becoming increasingly important, and various control charts have been developed to monitor different types of quality characteristics in this area. In general, the construction of conventional control charts to monitor multivariate processes in a high-dimensional setting has some statistical limitations and leads to misleading interactions. As a result, novel control charting techniques have recently been suggested to ameliorate the efficiency of monitoring schemes under high-dimensional data streams. This paper undertakes a conceptual classification structure using content analysis to classify the existing literature in this context for the period 2004–2024. Furthermore, on the basis of 72 selected papers, the research gaps are identified and some directions to stimulate potential for future research are suggested.

Trial sequential analysis in anesthesiology and intensive care: a systematic review

Introduction: The reliability of statistical analysis and its results is becoming a focal point in scientific discourse, overshadowing concerns about statistical significance. Systematic reviews and meta-analyses are characterized by a range of serious statistical and methodological limitations, including insufficient statistical power, clinical and statistical heterogeneity, systematic errors, potential low quality of included examinations, limited data availability, and publication bias. In this context, “Trial Sequential Analysis” (TSA) emerges as a promising methodological tool used in combination with meta-analyses to enhance the precision and robustness of scientific conclusions. Objective: The study is aimed to demonstrating potential scenarios for the using TSA and its role in improving the reliability of scientific conclusions through meta-analyses published in Russian journals on anesthesiology-resuscitation, as well as the authors’ own meta-analyses.Methods: This study is a systematic review without a detailed description of the case studies. The search for meta-analyses was conducted using the keywords “meta-analysis”, “mortality”, “delirium”, “sedation”. Two researchers independently reviewed the titles and abstracts of the selected meta-analyses and extracted data. TSA was conducted for all meta-analyses using “TSA software” (Copenhagen, Denmark). Results: 6 studies were considered, and 7 TSA evaluations were performed. Six out of the seven TSA evaluations confirmed the conclusions of the pertinent meta-analyses. In one TSA, the cumulative Z-line did not reach the required sample size, highlighting the need for additional studies to definitively confirm the result. In two TSAs, the sample size in the meta-analysis was insufficient to draw definitive conclusions. The application of TSA in these cases could provide more substantiated and cautious conclusions.Conclusion: This study explores various scenarios for the application of TSA in meta-analyses published in both Russian and international journals in the fields of anesthesiology and resuscitation. The significance of TSA for enhancing the validity and clinical relevance of research findings was substantiated. Received 3 February 2024. Revised 18 March 2024. Accepted 20 March 2024. Funding: The study did not have sponsorship. Conflict of interest: The authors declare no conflict of interest. Contribution of the authorsConception and study design: M.Ya. YadgarovData collection and analysis: M.Ya. Yadgarov, L.B. Berikashvili, P.A. PolyakovStatistical analysis: M.Ya. Yadgarov, P.A. PolyakovDrafting the article: M.Ya. Yadgarov, L.B. Berikashvili, K.K. Kadantseva, A.V. Smirnova, I.V. Kuznetsov, A.A. YakovlevCritical revision of the article: V.V. LikhvantsevFinal approval of the version to be published: M.Ya. Yadgarov, L.B. Berikashvili, P.A. Polyakov, K.K. Kadantseva, A.V. Smirnova, I.V. Kuznetsov, A.A. Yakovlev, V.V. Likhvantsev

Linear convergence of decentralized estimation for statistical estimation using gradient method

There has been a growing interest in solving consensus optimization problems in a multi-agent system. It is known that there is an exactness-speed dilemma for decentralized optimization for the naive gradient method, where either we have fast linear convergence to a neighborhood of size O(η) using a fixed constant step size η, or slow convergence to the exact optimizer with a decreasing step size. We reinvestigate a special case of this problem from a statistical point of view, where each agent’s local function is a sample estimate of a common population function. It is shown that when considering the convergence to the population target instead of to the exact consensual solution, the convergence is linear up to the statistical accuracy of the problem, using a step size only depending on the network characteristics instead of the sample size N. The objective function is not required to be strongly convex (but their statistical limit is), which allows for example the absolute deviation function which is used in median regression. Such results are in stark contrast to the optimization literature, where strong convexity of the objective function is necessary for linear convergence.

Structured collaborative sparse dictionary learning for monitoring of multimode processes

In this paper, a novel structured collaborative sparse dictionary learning approach is proposed to improve the monitoring performance of discriminative dictionary learning for multimode processes. The mode discriminability and data reconstruction are first balanced by decomposing the dictionary coefficients into between- and within-class parts and introducing a within-class self-expression regularization term. A weight vector of between-class coefficients is subsequently exploited for accurate mode identification of data that falls into the overlapping regions between different class distributions. Moreover, in order to pinpoint the fault variables, a scalable fault isolation method is developed which imposes a constraint of statistical control limit and introduces the ℓ1/ℓ2,0-structured sparsity regularization terms. The mode identification capability of the proposed method is proved theoretically by Theorems 1 and 2, and a lower-bound magnitude is provided by Theorem 3 for fault isolation. Finally, extensive experiments conducted in the numerical and industrial process demonstrate that our proposed method outperforms some state-of-the-art methods.

MODERN APPROACHES TO FORECASTING CORPORATE INCOME TAX REVENUES IN THE RUSSIAN FEDERATION

The purpose of the study. The purpose of the study is to determine the optimal model for forecasting corporate income tax revenues. The subject of the study is modern methods of forecasting income tax revenues to the budgets of the subjects of the Russian Federation, based on mathematical and statistical, regression and autoregressive models with a moving average. The article identifies the advantages and disadvantages of existing approaches to forecasting tax revenues, as well as defines the conditions for the applicability of the considered models in conditions of high dynamics of changes in market conditions and macroeconomic uncertainty. The results of the study are presented, which make it possible to establish the optimal methodology for forecasting corporate income tax receipts, taking into account the statistical limitations of the planning horizon, as well as to identify a forecasting mechanism available for use in long-term planning. Conclusions. Based on the results of the study, it was concluded that it is advisable to combine tax forecasting models to achieve optimal predictive and analytical indicators, as well as to apply this approach only within the framework of short-term planning. The final forecast indicator can be identified on the basis of expert assessments, the average value of the indicators obtained, or through a combination of these approaches: an expert assessment of the weighting coefficients to calculate the weighted average value of the forecast indicator.

A Perspective on Plasticity, Dissipation and the Second Law of Thermodynamics

Abstract The requirement of a non-negative dissipation rate for all possible deformation histories is generally imposed on plastic constitutive relations. This is a constraint analogous to the Coleman–Noll [Coleman, B. D., and Noll, W., 1964, “The Thermodynamics of Elastic Materials With Heat Conduction and Viscosity,” Arch. Ration. Mech. Anal., 13, pp. 167–178. 10.1007/BF01262690] postulate that the Clausius–Duhem inequality needs to be satisfied for all possible deformation histories. The physical basis for the Clausius–Duhem inequality is as a statistical limit for a large number of discrete events for a long time and is not a fundamental physical requirement for small systems for a short time. The relation between the requirement of a non-negative dissipation rate and the Clausius–Duhem inequality is considered. The consequences of imposing a non-negative dissipation rate for all possible deformation histories are illustrated for: (i) a single crystal plasticity framework that accounts for elastic lattice curvature changes as well as elastic lattice straining and (ii) for discrete defect theories of plasticity, with attention specifically on discrete dislocation plasticity for crystalline solids and discrete shear transformation zone (STZ) plasticity for amorphous solids. Possible less restrictive conditions on the evolution of dissipation in plasticity formulations are considered as are implications for stability. The focus is on open questions and issues.

Statistical limit superior and Statistical limit inferior in non-Archimedean L -fuzzy normed spaces

The purpose of this article is to study the notion of statistical limit superior(SLS) and statistical limit inferior(SLI) in non-Archimedean(NA) L -fuzzy normed spaces( L -FNS). The concept of SLS and SLI is examined and extended to SLS and SLI in NA L -FNS. Moreover, the analogue of some results between SLS and SLI over NA L -FNS have been discussed. And also, it is proved that a bounded sequence is statistically convergent over NA L -FNS. Throughout this article, K denotes a complete, non-trivially valued, non-Archimedean fields(NAF).