Estimated Regression Coefficients Research Articles

Bayesian regression modeling and inference of energy efficiency data: the effect of collinearity and sensitivity analysis

The majority of research predicted heating demand using linear regression models, but they did not give current building features enough context. Model problems such as Multicollinearity need to be checked and appropriate features must be chosen based on their significance to produce accurate load predictions and inferences. Numerous building energy efficiency features correlate with each other and with heating load in the energy efficiency dataset. The standard Ordinary Least Square regression has a problem when the dataset shows Multicollinearity. Bayesian supervised machine learning is a popular method for parameter estimation and inference when frequentist statistical assumptions fail. The prediction of the heating load as the energy efficiency output with Bayesian inference in multiple regression with a collinearity problem needs careful data analysis. The parameter estimates and hypothesis tests were significantly impacted by the Multicollinearity problem that occurred among the features in the building energy efficiency dataset. This study demonstrated several shrinkage and informative priors on likelihood in the Bayesian framework as alternative solutions or remedies to reduce the collinearity problem in multiple regression analysis. This manuscript tried to model the standard Ordinary Least Square regression and four distinct Bayesian regression models with several prior distributions using the Hamiltonian Monte Carlo algorithm in Bayesian Regression Modeling using Stan and the package used to fit linear models. Several model comparison and assessment methods were used to select the best-fit regression model for the dataset. The Bayesian regression model with weakly informative prior is the best-fitted model compared to the standard Ordinary Least Squares regression and other Bayesian regression models with shrinkage priors for collinear energy efficiency data. The numerical findings of collinearity were checked using variance inflation factor, estimates of regression coefficient and standard errors, and sensitivity of priors and likelihoods. It is suggested that applied research in science, engineering, agriculture, health, and other disciplines needs to check the Multicollinearity effect for regression modeling for better estimation and inference.

Estimating inter-regional trade: the radiation model versus the gravity model

The estimation of interregional trade flows within a country is confined to the context of multi-regional input–output models. Despite the remarkable relevance of these trade flows for regional incomes, most countries have almost no official data, with Korea being an exception used here. In addition to survey-based estimation methods, there are non-survey methods, some based on gravity models, which can be used to estimate inter-regional trade flows. This paper compares the gravity model with the novel radiation model. Preliminary results show that the accuracy of the radiation model’s estimation is superior to that of the gravity model, with the invaluable benefit of not needing to estimate regression coefficients, unlike the gravity model. This finding provides a generalisation of the radiation model for estimating interregional trade in contexts where there is no prior multi-regional input–output table without restrictions derived from coefficient estimation as in the gravity model.

Tensor-Based Statistical Learning Methods for Diagnosing Product Quality Defects in Multistage Manufacturing Processes

Multistage manufacturing processes with identical stages provide three-dimensional process data in which the first dimension represents the process (control/sensing) variable, the second is the stage, and the third is the measurement/sampling/data acquisition time point. Diagnosing quality faults in such processes often requires the simultaneous identification of crucial process variables and stages associated with product quality anomalies. Most existing diagnosis methods convert 3D data into a 2D matrix, resulting in loss of information and reduced diagnostic accuracy and stability. To address this challenge, we propose a penalized tensor regression model that regresses the product quality index against its 3D process data. For the estimation of high-dimensional regression coefficients with the limited amount of historical data, we apply the CANDECOMP/PARAFAC and Tucker decompositions to the coefficient tensor, which significantly reduces the number of parameters to be estimated. Based on the decompositions, a new regularization term is designed to enable the joint identification of critical process variables and stages. To estimate the parameters, we develop the block coordinate proximal descent algorithm and provide its convergence guarantee. Numerical studies demonstrate that the proposed methods can enhance diagnostic stability and on average improve the diagnostic accuracy by around 20% over existing benchmarks.

On the prediction of rare events when sampling from large data

When modeling rare events using logistic regression, independent samples of event occurrence (ones) and nonoccurrence (zeros) are commonly taken from large datasets in order to fit models efficiently. A deterministic offset may then be included in the model to compensate for that sampling method. We propose a more complex sampling approach using stratified sampling within the sets of ones and zeros to ensure that we may sample more zeros from strata of interest. This design may avoid situations in which a random sample of zeros fails to capture the range of a key covariate. We employ sampling weights along with stratum-specific intercepts to obtain unbiased estimates of the logistic regression coefficients (including the intercept) and their standard errors. We use simulation to show that this method provides unbiased parameter estimates comparable with those of maximum likelihood. We also illustrate an application of this method to wildland fire occurrence prediction in a study area in northwestern Ontario, Canada.

Temporal Recovery and Prognostic factors for Dysphagia Following Cardiovascular Surgery: Retrospective Analysis and Development of Predictive Score

ObjectiveDysphagia is a common complication associated with cardiovascular surgery (CVS). This study sought to better understand recovery timelines, identify risk factors, and create a prognostic model for oral intake restoration. MethodsThis retrospective study included 134 patients who had CVS between April 2022 and March 2024. We assessed swallowing ability through fiberoptic endoscopic evaluation of swallowing (FEES). We randomly divided the patients’ data into a training dataset and a test dataset in a ratio of 70/30 and Kaplan– Meier analyses and Cox regression were used to assess predictors of total oral intake. We also created a scoring system using the estimated regression coefficients. ResultsMost patients with CVS achieved total oral intake in 7-11 days after extubation. Over 168 h of intubation, the presence of penetration or aspiration, a poor FEES score (score >6), and perioperative complications were significant risk factors for delayed total oral intake. The predicting score was calculated by adding the scores for each individual factor, including FEES score, penetration/aspiration, and preoperative complications. Scores ranged 0–8, categorizing patients into 0–2, 3–5, and 6–8 groups, clearly demonstrating that the higher the predicting score, the longer the time to total oral intake in both the training and the test dataset. ConclusionsAll risk factors for unsuccessful or delayed total oral intake were intubation for more than a week, poor swallowing ability, and the presence of perioperative complications. The scoring system accurately predicts the restoration of oral intake following CVS.

Statistical Analysis of Self-Reported Health Conditions in Cohort Studies: Handling of Missing Onset Age

ObjectiveThis paper discusses methodological challenges in epidemiological association analysis of a time-to-event outcome and hypothesized risk factors, where age/time at the onset of the outcome may be missing in some cases, a condition commonly encountered when the outcome is self-reported. Study Design and SettingA cohort study with long-term follow-up for outcome ascer- tainment such as the Childhood Cancer Survivor Study (CCSS), a large cohort study of 5-year survivors of childhood cancer diagnosed in 1970-1999 in which occurrences and age at onset of various chronic health conditions (CHCs) are self-reported in surveys. Simple methods for handling missing onset age and their potential bias in the exposure-outcome association infer- ence are discussed. The interval-censored method is discussed as a remedy for handling this problem. The finite sample performance of these approaches is compared through Monte Carlo simulations. Examples from the CCSS include four CHCs (diabetes, myocardial infarction, osteoporosis/osteopenia, and growth hormone deficiency). ResultsThe interval-censored method is usable in practice using the standard statisti- cal software. The simulation study showed that the regression coefficient estimates from the ‘Interval censored’ method consistently displayed reduced bias and, in most cases, smaller stan- dard deviations, resulting in smaller mean square errors, compared to those from the simple approaches, regardless of the proportion of subjects with an event of interest, the proportion of missing onset age, and the sample size. ConclusionThe interval-censored method is a statistically valid and practical approach to the association analysis of self-reported time-to-event data when onset age may be missing. While the simpler approaches that force such data into complete data may enable the standard analytic methods to be applicable, there is considerable loss in both accuracy and precision relative to the interval-censored method.

Detection of earnings manipulations during the corporate life cycle in Central European countries

Research background: The growing trend of financial distress risk worldwide increases the need for effective tools to detect accounting manipulation by stakeholders (debt holders, shareholders, suppliers, and customers). However, the well-known models of accrual earnings management do not reflect the financial specificity of individual life cycle stages and their cash flow patterns. Purpose of the article: The study examines the impact of the corporate life cycle on the reliability of accrual earnings management models and identifies appropriate models for detecting accounting manipulation in Central European countries. Methods: Seven accrual earnings management models were used. They were evaluated based on five criteria: explanatory power (adjusted coefficient of determination), overall significance of the regression model, significance of the regression coefficients, predicted sign of the regression coefficients, and standard errors of the estimated regression coefficients. Finally, a comprehensive scoring model was used to determine the overall quality of the models examined. The study was conducted on a sample of more than 30,000 enterprises from four Central European countries (the Czech Republic, Hungary, Poland, and Slovakia), with data covering the period 2017–2021. The Dickinson non-sequential life cycle model was used to distinguish life cycle stages according to cash flow patterns. Findings & value added: The results of the study suggest that, firstly, discretionary accruals vary across countries as well as over the firm's life cycle; earnings manipulations have an inverted U-shape with mature firms tending to reduce their accounting profit. The reliability of the models examined was poor in the case of start-ups. This reflects the need to distinguish between life cycle stages in the detection of earnings manipulation.

Semi-parametric approach for modelling overdispersed count data with application to Industry 4.0

The paper deals with a test for the goodness-of-fit of a model for count data, in the framework of Generalized Linear Models. The motivating example concerns the study on the effectiveness of policy incentives for the adoption of 4.0 technologies by Small and Medium Enterprises. According to the literature, openness to Industry 4.0 should be measured in terms of the number of 4.0 technologies adopted, represented by a count variable. To investigate the effectiveness of public policy interventions to encourage the adoption of 4.0 technologies, we propose the application of a model for count data with a permutation ANOVA to test the goodness-of-fit and for the model selection. When the distribution of the response is neither Poisson nor Negative Binomial, and in the quite common situation in which the response variance is much greater than the mean, the classic Poisson regression and Negative Binomial regression are not valid. The proposed testing method is based on the combination of permutation tests on the significance of the regression coefficient estimates. The power behaviour of the proposed semi-parametric solution is investigated through a comparative Monte Carlo simulation study. The performance of such a method is compared to those of the two main parametric competitors. The application of the permutation test to an interesting case study is presented. The dataset is original, and related to a sample survey carried out in Italy, about the adoption of Industry 4.0 technologies by Italian enterprises.

Estimation and hypothesis test for varying coefficient single-index multiplicative models

ABSTRACT Estimation and hypothesis test for varying coefficient single-index multiplicative models are considered in this paper. To estimate an unknown single-index parameter, a profile product relative error estimation is proposed for the single-index parameter with a leave-one-component-out estimation method. A Wald-type test statistic is proposed to test a linear hypothesis test of the single-index. We employ the smoothly clipped absolute deviation penalty to simultaneously select variables and estimate regression coefficients. To study the model checking problem, we propose a variant of the integrated conditional moment test statistic by using a linear projection weighting function, and we also suggest a bootstrap procedure for calculating critical values. Simulation studies are conducted to demonstrate the performance of the proposed procedure and a real example is analysed for illustration.

Sustainable effects of a motor skill programme on physical activity levels in 7–8 years old children, in the Eastern Cape Province of South Africa

BackgroundDeteriorating global physical activity (PA) levels among children warrants new and sustainable approaches to increase PA levels. This study aimed to determine the immediate and sustainable influences of a 9-week movement program on the PA levels in 7 to 8-year-old school children in the Raymond Mhlaba Municipality in the Eastern Cape Province of South Africa.MethodsA randomized control trial including two groups (control group (CG) and intervention group (IG)), pre–post–retest (after six months of no intervention) design was used. Seventy school children, mean age 7.12 years (± 0.71) (n = 35 IG; n = 35 CG) participated in the study. A 9-week movement program was followed twice a week for 30 min during school hours. PA was measured for 7 consecutive days using a hip-mounted wGT3X-BT Actigraph accelerometer. The Test of Gross Motor Development-Third Edition (TGMD-3) was used to assess motor skills. Hierarchical Linear Modelling (HLM) was applied to analyze the data with time, sex, and group as predictors. Effect sizes were computed using Cohen’s d-cut points to assess the practical significance of changes over time. Estimated regression coefficients were also computed to determine the strength of the relationship between moderate-to-vigorous physical activity (MVPA) and fundamental movement skills (FMS).ResultsBefore the intervention, 60% of the IG met the 60 min of daily MVPA guideline, while light physical activity (LPA) per day was also higher than sedentary behavior (SB) in both groups. No immediate (p < 0.01) or sustainable (p < 0.01) increases in MVPA levels were found and no positive associations emerged between FMS and MVPA levels.ConclusionsThis intervention had little to no effect on children’s MVPA. More understanding of the activity behavior and interests of children is needed to improve their PA behavior through the content of movement programs. Strategies are also needed to communicate clear messages at a personalized but also parental level, focusing on enhancing health through regular PA, especially to promote PA in young children.

228 The effect of temperature-humidity index in different pregnancy stages on litter traits in Taiwan Landrace sows

Abstract Because of climate change, annual average temperatures are gradually rising. Hence, increasing heat stress can impair the litter traits of Taiwanese sows located in tropical and subtropical zones. However, some pregnancy stages may be impacted more by heat stress than other periods. Thus, the purpose of this study was to demonstrate the effect of temperature-humidity index (THI) in different pregnancy stages on total number born (TNB), number born alive (NBA), and stillborn rate (STB). Data were collected in two Taiwanese farms from 2008 to 2021 for TNB, NBA, and STB, while weather data were collected from the closest respective weather station. There were 4,247 and 6,812 pure line Landrace record from farm 1 and farm 2, respectively. Pregnancy stages included 28 d before mating (BEFORE), mating to 30 d of pregnancy (EARLY), 31 to 70 d of pregnancy (MIDDLE), and 71 d of pregnancy to farrowing (LATE). Average THI was calculated in each pregnancy stage and regression coefficients were estimated by linear mixed models within ASReml. A two-trait analysis was used to estimate regression coefficients and genetic correlations among TNB, NBA, and STB. Results showed THI impacted (P &amp;lt; 0.05) TNB and NBA BEFORE and EARLY, while THI influenced (P &amp;lt; 0.05) STB EARLY and MIDDLE. Regression coefficients for TNB were -0.0069 and -0.0228 for BEFORE and EARLY, respectively. For NBA, regression coefficients were 0.0106 and -0.0391 for BEFORE and EARLY, respectively. Regression coefficients for STB were -0.0651 and 0.1332 for EARLY and MIDDLE, respectively. Heritability estimates for TNB, NBA, and STB were 0.168, 0.113, and 0.069, respectively. Genetic correlations between TNB with NBA and STB were 0.913 and 0.539, respectively. While the genetic correlation between NBA and STB was 0.158. In conclusion, THI impacted BEFORE and EARLY stages for TNB and NBA, while the stages of EARLY and MIDDLE were impacted by THI for STB. Hence, control of the environments in these specific stages for sows could improve the performance of litter traits.

A prospective study of surgeons’ workloads and associated factors in real-world practice

New technologies such as laparoscopic and robotic surgery are spreading, and there is a demand for physicians to keep up with novel methods. In contrast to the recent focus on healthcare professional burnout, the mental and physical costs during surgery are not well-understood. We aimed to quantify surgeons’ workloads in daily urological surgical practice and clarify potential background factors associated with such workloads. Urologists in Hokkaido, Japan, were invited to this study. Between December 2020 and December 2021, participants repeatedly reported workloads, which were assessed using the National Aeronautics and Space Administration-Task Load Index (NASA-TLX), after each surgery in conjunction with participants’ names, patients’ backgrounds, their roles (independent operator, operator under supervision, instructor, and 1st or 2nd assistant), and surgical outcomes, via SurveyMonkey®. Because of the heterogeneity among individuals, a linear mixed-effects model was utilized to analyze factors associated with NASA-TLX, calculating the parameter estimates (PE) of regression coefficients for each factor and their 95% confidence interval (CI). Sixty-five urologists (5 women) joined the study, and 2169 data were collected within 7 days after surgeries. A linear mixed-effects model revealed that female surgeons (PE + 15.56, 95% CI 2.36–28.77), urgent/emergency surgery (PE + 6.65, 95% CI 4.59–8.70), intraoperative complications (PE + 9.26, 95% CI 6.76–11.76), and near-miss incidents (PE + 3.81, 95% CI 2.27–5.36) were associated with higher workloads. Regarding the surgeons’ role, operator under supervision (PE + 12.46, 95% CI 9.86–15.06) showed the highest workloads. Surgeons’ workloads decreased as the number of previous cases of the same procedure increased. Surgeons’ workloads were associated with various factors. Given that the highest workloads were for operators under supervision, instructors should be aware of trainees’ high workloads and devise appropriate instructional interventions.

A principal-weighted penalized regression model and its application in economic modeling

This paper introduces a novel Principal-Weighted Penalized (PWP) regression model, designed for dimensionality reduction in large datasets without sacrificing essential information. This new model retains the favorable features of the principal component analysis (PCA) technique and penalized regression models. It weighs the variables in a large data set based on their contributions to principal components identified by PCA, enhancing its capacity to uncover crucial hidden variables. The PWP model also efficiently performs variable selection and estimates regression coefficients through regularization. An application of the proposed model on high-dimensional economic data is studied. The results of comparative studies in simulations and a real example in economic modeling demonstrate its superior fitting and predictive abilities. The resulting model excels in accuracy and interpretability, outperforming existing methods.

The finite mixture model for the tails of distribution: Monte Carlo experiment and empirical applications

AbstractThe finite mixture model estimates regression coefficients distinct in each of the different groups of the dataset endogenously determined by this estimator. In what follows the analysis is extended beyond the mean, estimating the model in the tails of the conditional distribution of the dependent variable within each group. While the clustering reduces the overall heterogeneity, since the model is estimated for groups of similar observations, the analysis in the tails uncovers within groups heterogeneity and/or skewness. By integrating the endogenously determined clustering with the quantile regression analysis within each group, enhances the finite mixture models and focuses on the tail behavior of the conditional distribution of the dependent variable. A Monte Carlo experiment and two empirical applications conclude the analysis. In the well‐known birthweight dataset, the finite mixture model identifies and computes the regression coefficients of different groups, each one with its own characteristics, both at the mean and in the tails. In the family expenditure data, the analysis of within and between groups heterogeneity provides interesting economic insights on price elasticities. The analysis in classes proves to be more efficient than the model estimated without clustering. By extending the finite mixture approach to the tails provides a more accurate investigation of the data, introducing a robust tool to unveil sources of within groups heterogeneity and asymmetry otherwise left undetected. It improves efficiency and explanatory power with respect to the standard OLS‐based FMM.

On the effect of confounding in linear regression models: an approach based on the theory of quadratic forms

In the last two decades, significant research efforts have been dedicated to addressing the issue of spatial confounding in linear regression models. Confounding occurs when the relationship between the covariate and the response variable is influenced by an unmeasured confounder associated with both. This results in biased estimators for the regression coefficients reduced efficiency, and misleading interpretations. This article aims to understand how confounding relates to the parameters of the data generating process. The sampling properties of the regression coefficient estimator are derived as ratios of dependent quadratic forms in Gaussian random variables: this allows us to obtain exact expressions for the marginal bias and variance of the estimator, that were not obtained in previous studies. Moreover, we provide an approximate measure of the marginal bias that gives insights of the main determinants of bias. Applications in the framework of geostatistical and areal data modeling are presented. Particular attention is devoted to the difference between smoothness and variability of random vectors involved in the data generating process. Results indicate that marginal covariance between the covariate and the confounder, along with marginal variability of the covariate, play the most relevant role in determining the magnitude of confounding, as measured by the bias.

Variable Selection for Panel Data Linear Regression Models with Fixed Effects

This paper introduces a robust variable selection mechanism for fixed effect panel data models by integrating compound quantile regression with the adjusted MIXED penalty method. Initially, forward orthogonal deviation transformation is employed to eliminate the influence of fixed effects. Subsequently, the MIXED penalty is utilized to construct a penalized compound quantile regression objective function, facilitating simultaneous estimation of regression coefficients and variable selection. This method not only effectively eliminates the interference of fixed effects but also demonstrates outstanding robustness. Its performance with limited sample sizes was validated through simulation studies, and its practical value was illustrated through application in real data analysis.

Calibration Indonesian-Numerical Weather Prediction using Geostatistical Output Perturbation

Indonesian-Numerical Weather Prediction (INA-NWP) is a numerical-based weather forecast method that has been developed by the Meteorology, Climatology and Geophysics Agency. However, the forecast is still unable to produce accurate weather forecasts. Geostatistical Output Perturbation (GOP) is a weather forecast method derived from only one deterministic output. GOP takes into consideration the spatial correlation among multiple locations simultaneously. GOP is capable to identify spatial dependency patterns that are associated with error models. This study aims to obtain calibrated forecasts for daily maximum and minimum temperature variables using GOP at 10 meteorological stations in Surabaya and surrounding areas. The stages in performing temperature forecasts using GOP are obtaining regression coefficient estimators, then calculating empirical semivariograms and estimating spatial parameters. Based on several weather forecast indicators, such as RMSE and CRPS, GOP is better than INA-NWP in terms of precision and accuracy.

A Markov Chain Genetic Algorithm Approach for Non-Parametric Posterior Distribution Sampling of Regression Parameters

This paper proposes a genetic algorithm-based Markov Chain approach that can be used for non-parametric estimation of regression coefficients and their statistical confidence bounds. The proposed approach can generate samples from an unknown probability density function if a formal functional form of its likelihood is known. The approach is tested in the non-parametric estimation of regression coefficients, where the least-square minimizing function is considered the maximum likelihood of a multivariate distribution. This approach has an advantage over traditional Markov Chain Monte Carlo methods because it is proven to converge and generate unbiased samples computationally efficiently.

Poisson-modification of the Quasi Lindley optimal generalized regression estimator

In this paper, we propose an optimal generalized regression estimator based on the sample information for the Poisson-Modification of the Quasi Lindley (PMQL) regression model namely the PMQL optimal generalized regression estimator. Asymptotic properties of the proposed estimator are obtained. Further, the conditions of superiority over some other regression coefficient estimators for the PMQL regression model are also derived based on the mean square error criterion. In order to compare the performance of the proposed estimator over some other PMQL regression coefficient estimators in the scalar mean square error sense, a Monte Carlo simulation study is designed and results are discussed. Finally, a simulated data example and a real-world example are taken to show the potentiality of the new regression model based on the proposed estimator and show the consistency of the theoretical and simulation results, respectively.

Omics feature selection with the extended SIS R package: identification of a body mass index epigenetic multimarker in the Strong Heart Study.

The statistical analysis of omics data poses a great computational challenge given their ultra-high-dimensional nature and frequent between-features correlation. In this work, we extended the iterative sure independence screening (ISIS) algorithm by pairing ISIS with elastic-net (Enet) and 2 versions of adaptive elastic-net (adaptive elastic-net (AEnet) and multistep adaptive elastic-net (MSAEnet)) to efficiently improve feature selection and effect estimation in omics research. We subsequently used genome-wide human blood DNA methylation data from American Indian participants in the Strong Heart Study (n=2235 participants; measured in 1989-1991) to compare the performance (predictive accuracy, coefficient estimation, and computational efficiency) of ISIS-paired regularization methods with that of a bayesian shrinkage and traditional linear regression to identify an epigenomic multimarker of body mass index (BMI). ISIS-AEnet outperformed the other methods in prediction. In biological pathway enrichment analysis of genes annotated to BMI-related differentially methylated positions, ISIS-AEnet captured most of the enriched pathways in common for at least 2 of all the evaluated methods. ISIS-AEnet can favor biological discovery because it identifies the most robust biological pathways while achieving an optimal balance between bias and efficient feature selection. In the extended SIS R package, we also implemented ISIS paired with Cox and logistic regression for time-to-event and binary endpoints, respectively, and a bootstrap approach for the estimation of regression coefficients.