Robust Regression Model Research Articles

Disaggregating air, water and renewable energy disclosures in developing economies: the role of regulatory impact and board characteristics

PurposeThis study examines how board characteristics influence air, water and renewable energy (AWR) disclosures in an emerging economy. It argues for the necessity of separating these disclosures to address unique environmental impacts and stakeholder concerns.Design/methodology/approachUsing longitudinal data from environmentally sensitive firms (2014–2022), a disclosure index based on the Global Reporting Initiative (GRI) framework was developed to quantify AWR separately. To address potential statistical issues such as endogeneity and selection bias, the analysis employed a set of robust regression models, including the industry fixed effects (FE) model, a lagged model and a two-stage least squares (2SLS) model.FindingsBoard size and audit committees positively influence all AWR disclosures, while foreign directors significantly impact air and renewable energy disclosures. Board meetings negatively affect water disclosures. Surprisingly, board independence shows no significant impact, and gender diversity has no notable relationship. Post-amendment, firms increased AWR disclosures, though participation remains limited.Research limitations/implicationsGrounded in legitimacy theory, this study contributes to the literature by demonstrating how separating the unique characteristics of AWR disclosures offers stakeholders more precise insights into how firms manage specific environmental concerns. The findings are based on data from listed firms in Bangladesh and may not be generalisable to unlisted firms or other regions.Practical implicationsThe study emphasises the importance of distinct AWR reporting, offering valuable insights for regulators and corporate boards to improve transparency and sustainability practices.Social implicationsSeparating AWR disclosures provides stakeholders with clearer assessments of firms' environmental performance, promoting accountability and informed decision-making.Originality/valueThis study uniquely emphasises the need for disaggregating air, water and renewable energy disclosures in emerging economies. By focussing on each environmental issue separately, the research highlights how distinct disclosures offer clearer insights into how firms address specific environmental challenges, such as air pollution, water management and the transition to renewable energy sources. This disaggregation is essential for stakeholders – particularly regulators, investors and policymakers – to assess and respond to firms' sustainability efforts accurately.

Does Intellectual Capital Influence Firm's Financial Performance in Bangladesh?

This paper aimed to find the effects of intellectual capital and its components on firms’ financial performance in Bangladesh. A sample of 100 firms comprised of 48 manufacturing, 21 services, and 31 banking companies was studied for five years from 2017 to 2021, setting 500 firm-year observations. Data was collected purposively from secondary sources, such as annual reports of sampled companies. The robust fixed-effect regression model using STATA 14.2 software was applied to test the hypothesis due to autocorrelation and heteroscedasticity problems. The regression results documented that overall IC, human capital efficiency (HCE), and capital employed efficiency (CEE) positively and significantly enhanced the ROA and ROE of the firms. However, structural capital efficiency (SCE) and relational capital efficiency (RCE) negatively and insignificantly influenced the same. The study contributes to resource-based theory using econometric methods to extend the samples of both financial and non-financial companies. This study helps investors, managers, policy-makers, governments, and accounting regulatory bodies regarding the utilization pattern of invisible and tangible resources in Bangladeshi firms.

Abstract 4122679: Associations Between Recent and Cumulative Lifetime Cannabis Use and Sleep Health in Middle-Aged Adults: 35-Year Insights from the Coronary Artery Risk Development in Young Adults (CARDIA) and CARDIA Sleep Studies

Background: Cannabis use has become common, but its effects on sleep health—an important contributor to cardiovascular health—remain poorly understood. We investigated associations between recent and cumulative lifetime cannabis use and objective and subjective sleep measures in middle-aged adults. Method: We assessed recent cannabis use (last 30 days) repeatedly from young adulthood (age 18-30 at baseline in 1985-86) to middle age (after 35 years) in the longitudinal Coronary Artery Risk Development in Young Adults (CARDIA) study. Cumulative lifetime cannabis use (1 year = 365 days of use) was estimated by summing days of use during follow-up. At Year 35 (Y35), we obtained objective sleep measures (duration, efficiency, fragmentation, wake after sleep onset [WASO]) via 7-day wrist actigraphy from CARDIA Sleep Ancillary Study II. Subjective sleep quality was measured using the Pittsburgh Sleep Quality Index (PSQI). Robust regression models adjusted for Y35 age, sex, race, education, cumulative cigarette smoking, alcohol use, other substance use, physical activity, body mass index, sleep medication, and comorbidities. We also examined race, sex, and cigarette smoking status differences in associations. Result: Of 571 participants, mean age was 61.9 years; 65% were female, and 35% Black. Means (SD) for sleep duration, efficiency, fragmentation, WASO, and PSQI were 6.9 (1.1) hours, 90.5 (4.2)%, 20.6 (7.6)%, 43.5 (20.9) minutes, and 6.1 (3.6) scores, respectively. At Y35, 13.8% reported recent cannabis use, 56% reported former use, mean (SD) cumulative use was 1.6 (4.4) years. With adjustment, recent cannabis use was associated with longer sleep duration, lower efficiency, and higher WASO; former cannabis use was associated with longer sleep duration (Table). Compared to those with cumulative cannabis exposure years in the first quartile, fourth quartile users had longer sleep duration, lower efficiency, and higher WASO (β/p-value: 0.3/0.01, -1.2/0.01, and 8.9/<.001, respectively). PSQI was not associated with recent or cumulative cannabis use. Associations were similar across race, sex, and cigarette smoking status. Conclusion: Recent and cumulative lifetime cannabis use is associated with longer sleep duration and poorer sleep quality in objective, but not subjective, measures. Further longitudinal studies are needed to identify interventions aimed at improving sleep health and promoting cardiovascular health in cannabis users.

Prospective association between the Cardiovascular Health Diet Index and subclinical atherosclerosis: the ELSA-Brasil cohort study.

The Cardiovascular Health Diet Index (CHDI) is a diet quality score based on the dietary guidelines of the American Heart Association for cardiovascular health, but with some adaptations, such as red meat, dairy, beans, and ultra-processed foods in its components. The CHDI has shown good relative validity parameters, however, its association with health outcomes is still unclear. Thus, our aim was to investigate the association between the CHDI score with subclinical atherosclerosis. Data from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) cohort were used. Subclinical atherosclerosis was assessed by measuring coronary artery calcification (CAC) at baseline (2008-2010) and second wave (2012-2014) and carotid intima-media thickness (cIMT) at baseline and at the third wave (2017-2019). The CHDI score (ranges from 0 to 110 points) was applied to dietary data obtained from a food frequency questionnaire at baseline. Poisson regression with robust variance, linear regression and linear mixed-effects models were used to evaluate the association of the CHDI score with CAC incidence (n 2,224), CAC progression (n 725), and changes in cIMT (n 7,341) over time, respectively. After a median 8-year follow-up period, a 10-point increase in the CHDI score was associated with a decrease in cIMT of 0.002 mm (95% CI -0.005; -0.001). No association was observed between the CHDI score and CAC incidence and progression after a 4-year follow-up period. Higher scores in the CHDI were prospectively associated with decreased subclinical atherosclerosis after an 8-year follow-up period.

Debt maturity and climate change exposure: does life cycle moderate?

PurposeIn this study, I explore how climate exposure influences debt maturity and whether the life cycle moderates this relationship.Design/methodology/approachThis study employs robust panel fixed-effects regression models on 32,249 firm-year observations over 2002–2022. The meticulous examination of the relationship between climate exposure and debt maturity and the use of IV regression models to address endogeneity ensure the validity and reliability of the findings. Exploiting an external shock to validate the results further enhances the robustness of the analysis.FindingsThe study reveals that climate change exposure significantly negatively affects debt maturity, a finding that holds for both climate-related opportunities and regulatory risks. While the life cycle impacts debt maturity, it exacerbates the negative effect for firms in the introductory and growth phases. This exacerbation is due to underinvestment problems and a lack of financial flexibility. These findings shed new light on the relationship between climate exposure and debt maturity, providing valuable insights for corporate finance professionals and strategic managers.Practical implicationsThe evidence presented in this study has strategic implications for corporate managers, investors, academia and other stakeholders. The insights on the choice or substitution of debt maturity during certain life cycle phases can provide valuable guidance to management, empowering them to make informed decisions.Originality/valueThe dynamic role of the corporate life cycle in the relationship between capital structure and climate exposure is a novel area of research. This paper significantly contributes to our understanding of the combined relationship of firm life cycle and climate exposure on debt maturity, offering a fresh perspective on an underexplored topic. This study’s unique approach and original findings will pique the interest of academic researchers and corporate finance professionals alike.

Progression of aortic calcification among Japanese in Japan and white and Japanese Americans: a prospective cohort study.

Continued low mortality from coronary heart disease in Japan, despite deleterious changes in traditional risk factors, remains unexplained. Since aortic calcification (AC) was an early predictor of cardiovascular mortality, we compared the progression and incidence of AC between Japanese in Japan, white Americans, and third-generation Japanese Americans in the ERA JUMP cohort. We examined whether higher blood levels of marine-derived n-3 fatty acids (FAs) in Japanese than in Americans accounted for the difference. Men (n=700) aged 40-49 years (252 Japanese in Japan, 238 white, and 210 Japanese Americans) were examined at baseline and 4-7 years later. AC was evaluated from the aortic arch to the iliac bifurcation with computed tomography and quantified by the Agatston method. Robust linear regression and linear mixed models were used to compare the progression of AC. Multivariable logistic regression models were fitted to compare the incidence of AC (AC≥50 at follow-up) among those with baseline AC<50. Japanese in Japan had a significantly slower progression of AC than white and Japanese Americans after adjusting for age, baseline AC, follow-up time, and traditional risk factors. White Americans had a significantly higher incidence of AC than Japanese in Japan (OR=4.61 [95%CI, 1.27-16.82]). Additional adjustment for blood levels of n-3 FAs accounted for the difference in AC incidence but not progression. Japanese in Japan had a significantly slower progression and lower incidence of AC than white Americans. High levels of marine-derived n-3 FAs in Japanese in Japan partly accounted for the difference in incidence.

The impact of auditor attributes and firm size on financial reporting timeliness of listed firms

This empirical study examines the impact of auditor attributes and firm size on financial reporting timeliness among listed firms in Nigeria. The study employs an ex-post facto type of research, with a quantitative design covering a ten-year period (2013–2022). The sample size comprises sixty-six (66) non-financial firms listed on the Nigerian Exchange Group (NGX). Based on data extracted from the audited annual reports of the sampled sixty-six firms, the robust regression model results reveal that joint audits contributed considerably to shorter financial reporting lags, underscoring the value of collaborative audit efforts in streamlining the audit process. Audit fees maintained a positive significant effect on the reporting lag of listed Nigerian firms. However, audit switch, client firm size, audit opinion, and audit firm size all maintained insignificant effects on the financial reporting timeliness of the Nigerian listed firms investigated. Therefore, the study recommends that listed firms should rather opt for affordable joint audits due to their efficiency in streamlining the audit process. Equally, the study recommends that listed firms should maintain long-term relationships with auditors to leverage increased familiarity, yet remain cautious of likely complacency and breach of auditing ethical guidelines that can arise from prolonged engagements.

Machine Learning Approach to Rapidly Evaluate Curling of Concrete Pavement

This paper focuses on the methodology for evaluating the degree of total curling in concrete pavement using machine learning. Deflection induced by falling weight deflectometer (FWD) testing is known as a direct correlation to total curling including built-in and daily curling. However, deflection measurement in the in-service road is also affected by others, such as environmental conditions, pavement geometry, subgrade stiffness, and mixture design. Thus, it is challenging to determine the level of curling from FWD data due to the complexity of influencing parameters. To navigate this complexity, prominent machine learning models are exploited to identify a non-linear relationship between curling and FWD deflections. A finite-element simulation of FWD is conducted to generate a vast data set, and a robust regression model is trained to estimate the total effective temperature difference (TETD) to quantify the effects of curling. Since input parameters for testing pavements can be measurable in the field, curling from TETD can be readily obtained using the proposed methodology. Comparative simulations highlight that the proposed models, with an MAE less than 0.5 °C significantly outperform the multiple regression performance, which registers an MAE of 3.45 °C in TETD, particularly in offering cost-effective and noise-tolerant prediction.

Theoretical investigations on the purification of petroleum using desulfurization process: Analysis and optimization of process

This paper introduces a comprehensive approach for predicting chemical concentrations (C) of sulfur compound in a two-dimensional space (x, y) by numerical solution of mass transfer equation and integration of machine learning. Data of training/validation are obtained from mass transfer modeling on adsorption separation utilizing porous adsorbent for petroleum desulfurization. Mass transfer modeling data were utilized for machine learning models which included three different regression models, namely Support Vector Machine (SVM), Decision Tree Regression (DT), and Multilayer Perceptron (MLP). The hyper-parameters of these base models were optimized using the Grey Wolf Optimization (GWO) algorithm. The ensemble models, denoted as ADA-DT, ADA-MLP, and ADA-SVM were assessed based on key performance metrics, including R2, MAE, RMSE, and MAPE. Results demonstrated the efficacy of the ensemble models in capturing complex relationships within the dataset. ADA-DT obtained an exceptional R2 score of 0.99031, highlighting its outstanding predictive accuracy. Similarly, ADA-MLP and ADA-SVM showed great accuracy, achieving R2 of 0.88272 and 0.96842, respectively. In this study, we uncover valuable insights into the application of ensemble methods and hybridized optimization methods for accurate and robust regression modeling in chemical concentration prediction scenarios applicable for petroleum engineering.

Fuzzy robust regression based on exponential-type kernel functions

The least squares method is a frequently used technique in fuzzy regression analysis. However, it is highly sensitive to outliers in the dataset. To address this challenge, we propose a novel robust fuzzy regression model based on exponential-type kernel functions. This approach effectively mitigates the influence of poorly fitted observations on the predicted results by reducing their weights. Furthermore, we use the gh-transformation to guarantee the nonnegativity of the spreads of the predicted response variable. In order to evaluate the performance of our method, a simulation study and three real data sets were considered. The experimental results demonstrate that the proposed method outperforms several popular robust methods in almost all cases. Furthermore, a sensitivity analysis of the estimated parameters provides further evidence of the superior efficiency of the proposed method.

Environmental impacts of the billion tree Tsunami project in Khyber Pakhtunkhwa on the dynamics of Agro-Meteorological Droughts

Changes in forest cover are closely associated with the variability in meteorological and hydrological variables. Therefore, this study delves into investigating how forest cover changes impact the environment (i.e., hydro-meteorological variables, including precipitation, streamflow, relative humidity (RH), evapotranspiration (ET), and temperature) using Trend Projection (TP) methods during 1980–2019. The study is carried out in the Khyber Pakhtunkhwa (KP) province of Pakistan, which witnessed deforestation between 1980 and 2010 followed by afforestation (through billion tree tsunami project, BTTP) initiated in 2014. A new drought index, named as agro-meteorological drought index (AMDI), is developed in this study using the remotely sensed data to analyze the impact of forest cover on drought severity. The robust least square regression (RLSR) model is used to regress the normalized difference vegetation index (NDVI) with AMDI at various time scales to investigate the impact of forest cover on drought severity. The RLSR and paired t-test are used to quantify the impact of forest cover and BTTP, in particular, on the environment. Land use maps prepared for KP province over a span of the past four decades revealed significant deforestation during 1985–2005, transitioning gradually to afforestation in the past decade. Results indicated a decline in streamflow throughout different seasons with an increase in forest cover, particularly during the period of afforestation (i.e., 2015–2019). The precipitation, RH (maximum/minimum), and ET displayed an increasing trend over time, whereas a decrease trend is observed in Tmax/Tmin and streamflow at most of the stations. The trend analyses depicted a significant change before and after the BTTP. The paired t-test results revealed that BTTP has statistically significant impact on the environmental variables. Furthermore, the time series plots of AMDI at different time scales indicated that drought events were frequent and severe prior to 2003, whereas significant decrease in both the frequency and severity of drought was observed in the last decade (2010–2019). The RLSR results at pixel scales demonstrated the crucial role of forest covers in alleviating both the frequency and severity of drought events. The elasticities revealed that increase in the forest cover resulted in substantial increase/decrease in each hydro-meteorological variable. Overall, the results highlighted a positive and statistically significant impact of forest cover (i.e., BTTP) on both the environment and drought variability.

Tailoring the Nutritional Composition of Italian Foods to the US Nutrition5k Dataset for Food Image Recognition: Challenges and a Comparative Analysis

Background: Training of machine learning algorithms on dish images collected in other countries requires possible sources of systematic discrepancies, including country-specific food composition databases (FCDBs), to be tackled. The US Nutrition5k project provides for ~5000 dish images and related dish- and ingredient-level information on mass, energy, and macronutrients from the US FCDB. The aim of this study is to (1) identify challenges/solutions in linking the nutritional composition of Italian foods with food images from Nutrition5k and (2) assess potential differences in nutrient content estimated across the Italian and US FCDBs and their determinants. Methods: After food matching, expert data curation, and handling of missing values, dish-level ingredients from Nutrition5k were integrated with the Italian-FCDB-specific nutritional composition (86 components); dish-specific nutrient content was calculated by summing the corresponding ingredient-specific nutritional values. Measures of agreement/difference were calculated between Italian- and US-FCDB-specific content of energy and macronutrients. Potential determinants of identified differences were investigated with multiple robust regression models. Results: Dishes showed a median mass of 145 g and included three ingredients in median. Energy, proteins, fats, and carbohydrates showed moderate-to-strong agreement between Italian- and US-FCDB-specific content; carbohydrates showed the worst performance, with the Italian FCDB providing smaller median values (median raw difference between the Italian and US FCDBs: −2.10 g). Regression models on dishes suggested a role for mass, number of ingredients, and presence of recreated recipes, alone or jointly with differential use of raw/cooked ingredients across the two FCDBs. Conclusions: In the era of machine learning approaches for food image recognition, manual data curation in the alignment of FCDBs is worth the effort.

Evaluation of supervised machine learning regression models for CFD-based surrogate modelling in indoor airflow field reconstruction

Fast and reliable prediction of indoor airflow distribution is critical for indoor environment control. While neural networks (NN), often interchangeably referred to as Back Propagation Neural Networks (BPNNs), are popular for airflow predictions, optimizing these models is challenging due to their ”black box” nature and complex network structures. This study explores alternative robust regression models, including decision-tree-based models (e.g., XGBoost, LightGBM, Random Forest) and Support Vector Regression (SVR), for predicting indoor airflow. Two BPNN structures were initially developed to evaluate feasibility of NN models. BPNN A was trained using airflow velocities from two inlets as input neurons to directly predict the airflow velocity distribution within the domain. BPNN B was trained additionally with spatial information, including space samples and boundary wall data. Higher-dimensional training structures of BPNN B were applied to decision tree-based models and SVR to assess their capability in predicting non-linear airflow patterns. Results indicated that BPNN A achieved the highest accuracy, while the inclusion of higher-dimensional data in BPNN B led to decreased accuracy. Among all decision-tree-based models, XGBoost demonstrated the greatest potential, achieving an R2 above 99.5% and predictive errors below 10%. XGBoost also outperformed both BPNN models in speed, being 15.78 times faster than BPNN A and 252 times faster than BPNN B. The interpretability of XGBoost was further explored by analysing feature importance, which helps identify the most influential input variables while predicting the airflow velocity. This analysis is expected to offer an enhanced understanding of boundary conditions leading to optimised indoor environment strategy.

Internet-based therapist-supported interpersonal psychotherapy for depression: A randomized controlled trial

BackgroundDepression is a common disorder for which there are several treatments options including different psychological treatments. The aim of this study was to investigate the effects of internet-based interpersonal psychotherapy (IPT) for symptoms of depression in randomized controlled trial. MethodsFollowing recruitment via advertisement a total of 113 participants with mild to moderate symptoms of depression were included and randomized to either a ten-week internet-based IPT with weekly therapist guidance or a waitlist control condition. The primary outcome was symptoms of depression measured weekly with the Montgomery Åsberg Depression Rating Scale (MADRS-S) and at pre- and post-treatment assessment with the Beck's Depression Inventory (BDI-II). Secondary outcomes were self-rated quality of life and symptoms of generalized anxiety disorder. We also measured therapeutic alliance and treatment credibility. Outcomes were evaluated with a latent growth curve model (for MADRS-S) and robust linear regression models (for the other measures). The trial was conducted during the Covid-19 pandemic in the spring of 2021. ResultsSignificant differences favoring the treatment group were found on three of the four outcomes: BDI-II, quality of life ratings, and ratings of generalized anxiety. Between-group effect sizes for these outcomes were moderate (BDI-II, quality of life) or small (generalized anxiety). The latent growth curve model did not indicate a significant difference on the weekly MADRS-S ratings. Exploratory analyses did not show an association between therapeutic alliance, treatment credibility and outcome. LimitationsMissing data at post-treatment was high in the treatment group (37 %), though the missingness was not significantly related to observed ratings at pre-treatment or estimated trajectories during the treatment. Few participants completed all modules. The Covid-19 pandemic situation may have affected both effects and dropout rates. ConclusionsInternet-based IPT can lead to significant improvements, though the reductions in symptoms of depression were not consistent across the two measures used. Completion rates and dropout patterns suggest a need for improved acceptability. Trial registrationThe trial was preregistered at Clinicaltrials.gov (Identifier: NCT04721678). Registered January 2021.https://clinicaltrials.gov/study/NCT04721678

Predictive Utility of the Multi-Process Action Control Framework for Self-Reported and Device-Measured Physical Activity Behavior of Adolescents.

Understanding the correlates of physical activity behavior is imperative for informing the development of interventions to address the low rates of physical activity guideline adherence among adolescents living in the United States. This cross-sectional study examined the predictive utility of the Multi-Process Action Control (M-PAC) framework for explaining self-reported and device-measured physical activity behavior among a Hispanic-majority sample of adolescents. A total of 1849 high school students (mean age = 16.0 ± 1.22 SD years; 52.3% women; 87.8% Hispanic) enrolled in one school district in south-central Texas completed a survey including instruments to assess M-PAC framework constructs (instrumental and affective attitudes, perceived capability and opportunity, behavioral regulation, habit, identity) and moderate-to-vigorous physical activity (MVPA) behavior. A subsample (n = 435) wore accelerometers for seven days. The results from robust linear regression models revealed role identity and habit were significant predictors of self-reported MVPA. Role identity was a significant predictor of accelerometer-derived daily MVPA and raw acceleration in the most active hour but not daily raw acceleration. The findings indicated reflexive processes are robust predictors of adolescent physical activity and should be the focus of interventions designed to promote adoption and maintenance of physical activity during this developmental life stage.

Evaluation of a hyperspectral image pipeline toward building a generalisation capable crop dry matter content prediction model

Hyperspectral imaging has proven to be a reliable technique for estimating dry matter, a common variable when considering the quality of the fresh produce. However, developing models capable of generalising across different crops is challenging. In this study, several pipelines were explored towards achieving a robust and accurate generic regression model were evaluated and the development of Automatic Relevance Determination (ARD) and Partial Least Squares (PLS) algorithms for fruit and vegetable dry matter estimation. The models were built using a VIS-NIR dataset that includes both fruit and vegetables, namely, apples, broccoli and leek (n = 779). The PLS regression model obtained Root Mean Square on Prediction (RMSEP) = 0.0137, outperforming ARD regression (RMSEP = 0.0140) on a 10x5-fold cross-validation protocol. The evaluated preprocessing techniques affect the two regression algorithms differently, with the best results achieved when the pipeline was used without feature extraction. Overall, the pipeline using either ARD or PLS regression shows strong performance and generalisation for Visible-Near Infrared (VIS-NIR)-based dry matter estimation across diverse fruits and vegetables.

Utilizing Additive Manufacturing for Economical Prosthetic Limb Prototyping: A Guide from Regression Modelling

This study investigated the development of a low-cost prosthetic limb using additive manufacturing techniques, guided by anthropometric data and regression modeling. Anthropometric measurements were collected from 250 healthy young adults to characterize limb dimensions. Statistical analysis revealed the mean, standard deviation, and range for right and left leg and elbow lengths. A regression model was formulated to estimate limb lengths based on these measurements. A robust regression model demonstrated a significant correlation (r = 0.993, r² = 0.986) between right limb measurements and corresponding left limb measurements, indicating that right leg and elbow lengths could accurately predict left side measurements. The derived regression equation (RLL = 2.253 + 0.959 LLL + 0.054 LEL) provides a practical tool for designing customized prosthetics. Utilizing 3D printing technology, these findings can enhance the production of tailored prosthetic limbs, improving comfort and functionality for users. The study recommends further validation across diverse populations, integration into clinical practices, and continued research and development supported by policy and funding initiatives. These steps will advance the accessibility and effectiveness of personalized prosthetic solutions.

The Robust Negative Binomial Regression Model on Under-five Mortality due to Pneumonia in the Province of East Java

Robust Negative Binomial regression model (RNBR) is a modelling method to overcome a problem if there are outliers and overdispersion in the data. Outliers are data points that are significantly different from other data. Outliers have a significant effect on modelling to the resulting model. Furthermore, overdispersion is indicated by the presence of too large values of Pearson statistics. In this study, the RNBR model was used to determine the factors of the toddler immune variable at post neonatal age that significantly influenced the number of under-five deaths caused by pneumonia in East Java Province. Based on the modelling obtained, it shows that the RNBR model provides more robust results in handling outlier and overdispersion problems. This can be seen from the AIC value of the RNBR model is smaller than the AIC of the Poisson regression model. In addition, and which are measures of the influence of outliers on the model, decreased from 1 for the Poisson regression model to around 0.42 for the RNBR model.

How enabling factors determine unmet healthcare needs? A panel data approach for countries

Health service need refers to the essential care required to achieve optimal health outcomes within resource constraints. When necessary services to address identified health issues are not received, unmet needs arise. This research focuses on the determinants of unmet healthcare needs across the 34 countries within the European region from 2011 to 2019, focusing on Andersen's Behavioral Model's enabling factors. We employed a static and robust panel regression model using Stata 14.0 software. Key determinants analyzed include GDP per capita, urbanization rate, and physicians per capita. Findings reveal that lower GDP per capita and lower urbanization rates are significantly correlated with higher levels of unmet healthcare needs, highlighting income level and geographical accessibility as critical factors. Additionally, a higher number of physicians per capita is associated with reduced unmet healthcare needs, indicating the importance of healthcare resources in addressing healthcare access gaps. These findings underscore the importance of targeted healthcare policies that address income level, improve healthcare accessibility, and enhance healthcare resource allocation to reduce unmet healthcare needs effectively. These findings equip policymakers and administrators with empirically grounded insights to comprehend the factors contributing to unmet healthcare needs and to develop policies aimed at addressing this challenge.

Инновации как фактор экономического роста: регрессионный анализ мировых экономик / Innovation as a Factor of Economic Growth: Regression Analysis of World Economies

This paper investigates the relationship between innovative development and macroeconomic indicator of GDP per capita in different countries. Using a robust regression model, we analyze how investment in innovation, research and development (R&amp;D) expenditures, and other innovation-related factors contribute to economic growth. Our results show a significant positive correlation between the innovation index and GDP per capita, suggesting that countries that prioritize innovation tend to achieve better economic performance. These findings emphasize the importance of fostering innovation to enhance economic prosperity and provide a roadmap for sustainable growth.