Robust Regression Research Articles

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.

Geographical Mapping of COVID-19 Testing Rates in St. Louis: Influence of the Socioeconomic Index and Race on Testing in the Early Phase and Peak of the COVID-19 Pandemic

Purpose: This study aimed to explore how race and the socioeconomic index (SES) of residential zip codes influenced COVID-19 testing rates in St. Louis during the early and peak phases of the pandemic. Method: An ecological study was conducted using COVID-19 testing data from March to November 2020 in St. Louis City, including 16,915 tests from 30 residential zip codes. Geographical mapping identified areas lacking testing, descriptive statistics characterized testing rates by race and SES, and robust linear regression modeled the association between testing rates and race, controlling for SES. Results: The overall testing rate was 303.97 per 10,000 population. Testing rates among Blacks, Whites, American Indians/Alaskan Natives, and Asians were 374.26, 247.77, 360.63, and 242.41 per 10,000, respectively. Higher rates were observed in low- and middle-SES zip codes compared with high-SES areas. Adjusted for SES, testing rates per 10,000 were higher by 125 (1.25%), 448 (4.48%), and 32 (0.32%) among Blacks, American Indians/Alaskan Natives, and Asians compared with Whites. Testing rates decreased by 56 (0.56%) per unit SES improvement. There was statistically significant SES finding and that of the difference between American Indians/Alaskan Natives and Whites. Geographical mapping showed higher rates in the inner core and lower rates in the periphery of St. Louis. Conclusions: Race and SES influence testing rates. Targeted interventions are needed in areas with low testing rates in St. Louis.

Navigation performance in glaucoma: virtual-reality-based assessment of path integration

Navigation is essential for moving between locations in our daily lives. We investigated the relationship between visual impairment in glaucoma and path-integration-based navigation. Fourteen glaucoma and 15 controls underwent ophthalmological examination (including visual acuity (logMAR), visual field sensitivity (MD: mean deviation from matched reference cohort), and peripapillary retinal nerve fiber layer (pRNFL)). Both groups navigated physically in virtual reality (VR) environments during daylight and dawn conditions. Briefly, the participants traversed a path marked by three targets, subsequently pointing back to the path’s origin. Outcome measures included (i) travel-time, (ii) pointing-time, and (iii) Euclidian-distance error between indicated and starting position. Robust linear regression was conducted between visual function outcomes of the better eye and VR outcome measures. Glaucoma patients showed increase in travel-time (by 8.2 ± 1.7 s; p = 0.002) and in pointing-time (by 5.3 ± 1.6 s; p = 0.016). Predictors were MD for all outcome measures (p < 0.01) and pRNFL for travel-time (p < 0.01). The results suggest that the effect of glaucoma on the elapsed time depends on disease progression, i.e. people with stronger visual impairment need more time. This uncertainty during everyday navigation tasks may adversely affect their quality of life.

Forecasting of meteorological drought using ensemble and machine learning models

This study highlights drought forecasting for understanding the semi-arid area in India, where drought phenomena play vital role in the irrigation, drinking water supplies, and sustaining the ecological with economic balance for every nation. Therefore, drought forecasting is important for the future drought planning based on the machine learning (ML) models. Hence, The Standardized Precipitation Index (SPI) at 3- and 6-month periods have been selected and used for future drought forecasting scenarios in area. The combinations of ten inputs SPI-1- and SPI-10 were used for predicting modeling for SPI-3 and SPI-6 timescales, that modeling developed based on the historical SPI datasets from 1989 to 2019 years. The SPI-3 and SPI-6 maximum and minimum values are shown SPI-3 (2.03 and -5.522) and SPI-6 (1.94 and -6.93). The SPI is a popular method for estimating the drought analysis and has been used everywhere at global level. The developed models have been compared with each other, with the best combination of input variables selected using subset regression models and sensitivity studies. After that, the active input parameters were used for forecasting of SPI-3 and SPI-6 values to understanding of drought in semi-arid area. The finest input variables combination have been used in the Ml models and established the novel five models such as robust linear regression, bagged trees, boosted trees, support vector regression (SVM-Linear), and Matern Gaussian Process Regression (Matern GPR) models. Such kind of models first time has been applied for the forecasting of future drought conditions. Whole models were fine and improved modeling by using hyperparameters tuning, bagging, and boosting models. Entire ML models’ accuracy was compared using different statistical metrics. Compared with five ML models accuracy, we have found that the Matern GPR model better accuracy than other ML models. The best model accuracy is R2 = 0.95 and 0.93, RMSE, MSE, MAE, MARE, and NSE values, respectively, for predicting SPI-3 and SPI-6 values in the area. Therefore, the Matern GPR model was identified as the finest ML algorithm for predicting SPI-3 and SPI-6 associated with other algorithms. This research demonstrates the Matern GPR model's efficacy in predicting multiscale SPI-3 and SPI-6 under climate variations. It can be helpful in soil and water resource conservation planning and management and understanding droughts in the entire basin areas of the country India.

Do the Characteristics of Startup Founders Matter for Funding Performance?

Introduction/Main Objectives: Based on the upper echelons theory, this research examines the role of startup founders' characteristics on startup funding performance (SFP). This study considers the founder's experience, education, and gender as the characteristics that positively affect the SFP. Background Problems: While academia and industry emphasize the importance of startup funding performance (SFP), the empirical evidence on its determinant has received less attention. Yet, upper echelons theory addresses the importance of a leader's characteristics in increasing the organization's performance. Therefore, the current study focuses on investigating the role of startup founders' characteristics in the SFP. Novelty: The current study proposes the founder's characteristics as the key factor for startup funding. Unlike previous studies, this study utilized a survey design to answer the research question. Additionally, this study is the pioneer for entrepreneurial finance studies in gauging the upper-echelons framework. Research Methods: This research collected information from 228 Indonesian Startups through a survey. The estimation model is estimated using robust cross-sectional OLS regression and logistic regression. For robustness purposes, this study tackles the endogeneity issue by using two-stage least squares (TSLS) and PLS-SEM. Finding/Results: The regression results (including the two-stage least square approach) reveal that education and gender play a significant role in SFP but not experience. Moreover, it shows that higher education would increase a startup's probability of having better funding. Meanwhile, male entrepreneurs have lower funding performance than women entrepreneurs. Conclusion: Our study surmises that the education of the founders is crucial for startup financing in Indonesia. It also shows that the experience of founders has a trivial effect. It supports the upper echelons theory, emphasizing the pivotal role of founders' characteristics in entrepreneurial finance. Policy-wise, focusing on education can enhance startup success while addressing gender disparities is crucial. However, limitations exist, suggesting the need for longitudinal studies and broader sample sizes. Future research could explore managerial abilities and cultural factors, offering fresh insights into the literature.

Travelers With Disabilities: How Online Information Search Reduces Uncertainty About Service Accessibility in Hospitality and Tourism

Despite the significant number of travelers with disabilities and their substantial contribution to the tourism and hospitality industry, there has been a lack of research exploring the necessary services required by travelers with disabilities during their travels. Therefore, this study aims to identify accessibility issues encountered by travelers with disabilities during their travels and to examine the impact of these issues on their behavior in reducing uncertainty within the online travel community. To achieve the research objectives, the present study performed an integrated approach that combines text analytics and robust regression analysis. The findings of this study provide guidelines for tourism and hospitality management to reduce the perceived uncertainty experienced by travelers with disabilities, thereby promoting the accessibility of tourism and hospitality services.

Working capital and performance: Europe’s challenge to crises

This paper explores the relationship between working capital management (WCM) strategies and business performance during the COVID-19 crisis. Data from 1,067 non-financial European firms from 33 countries distributed across 6 economic sectors in the period from 2015 to 2021, feed Ordinary Least Squares model (OLS), robust and quantile pooled regression. COVID-19 crisis seems to have negatively influenced only median-low-range performance firms in Europe, showing that crisis coming from productive capacity hibernation do not impact nor impose tough slow recovery, transversal to all firms, as occurred on other financial crises. As for WCM, only median-top-range performance firms had significant statistical evidence for negative impact, that become positive for median-range performance firms during COVID-19. For Eurozone firms, there was an additional positive influence of the WCM on performance during COVID-19. Median-range performance firms seems to benefit from conservative WCM strategies mitigating firms’ performance negative effect raised by COVID-19 crisis. Relation between COVID-19 and WCM seems to be sensitive to different levels of firms’ performance and to different business’ economic sector. Contrary to knowledge on WCM, performance and financial crises in Europe, for COVID-19 alike crises, performance benefits from higher cash conversion cycles and thus encourage conservative WCM strategies. Liquidity should not be pushed by adopting aggressive WCM strategies.

Joint Impact of Market Volatility and Cryptocurrency Holdings on Corporate Liquidity: A Comparative Analysis of Cryptocurrency Exchanges and Other Firms

This study examines the impact of market volatility and cryptocurrency holdings on corporate liquidity, with a particular focus on the differences between cryptocurrency exchanges and other businesses. The analysis is based on 181 firm-year observations from 2017 to 2022, using Bitcoin volatility, VIX, and VKOSPI as indicators of market volatility. Ordinary Least Squares (OLS) and robust regression analyses are employed to assess the relationships between these variables. It is first noted that, albeit insignificant, market volatility has a detrimental influence on company liquidity. The positive correlation for cryptocurrency exchanges, however, suggests that cryptocurrency exchanges could potentially leverage market volatility as a strategic advantage. Additionally, the study shows that cryptocurrency holdings enhance corporate liquidity, with a stronger association observed in cryptocurrency exchanges. The analysis also incorporates lagged variables to capture delayed effects, confirming that cryptocurrency holdings exert both immediate and delayed positive impacts on liquidity, likely due to effective strategic management practices within exchanges.

The drive for muscularity and disordered eating among heterosexual and sexual minority men in Israel.

In this study, we investigated the moderating effects of self-acceptance and perceived country-level acceptance of gender and sexual orientation on the relationship between the drive for muscularity and disordered eating, particularly among sexual minority and cisgender men. We hypothesized that high levels of acceptance, both individually and within the societal context, would mitigate the pressure to adhere to conventional masculine ideals, consequently lowering the likelihood of engaging in disordered eating behaviors. Seven hundred twenty-three Israeli men aged 18-68 participated in the study. The sample included heterosexual men (67.9%) and sexual minority men (30.1%), with a deliberate oversampling of sexual minority men to explore differences in the suggested model between heterosexual and sexual minority groups. The data were analyzed using hierarchical robust regression, examining interactions between the variables of interest. The results revealed significant associations between the drive for muscularity, disordered eating, and lower self-acceptance of sexual identity. Sexual minority men reported higher levels of disordered eating and drive for muscularity than did heterosexual men, along with lower levels of acceptance of gender and sexual orientation (by the individual and the country). The association between drive for muscularity and disordered eating was moderated by sexual orientation and self-acceptance of gender identity, with a stronger association among men with lower self-acceptance of their gender identity and sexual minority men. The findings underscore the significance of fostering individual and country-level acceptance of gender and sexual orientation differences to promote well-being and mitigate the risk of disordered eating among heterosexual and sexual minority men, particularly the latter. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

ORGANIZATIONAL CULTURE AND ITS INFLUENCE ON EMPLOYEE JOB SATISFACTION

This research paper examines the impact of organizational culture on employee job satisfaction. The study hypothesizes that elements such as pride in the organization, a positive culture, leadership inspiration, motivation, supervisor support, values alignment, a sense of belonging, values influence, open communication, informed decisions, and comfort in sharing ideas are significant predictors of job satisfaction. Utilizing a survey administered to 500 employees, responses were analyzed using multiple regression and robust regression techniques. The findings reveal that "Leadership Inspires," "Supervisor Support," and "Sense of Belonging" are statistically significant factors positively influencing job satisfaction. Descriptive statistics indicate a generally high level of job satisfaction among the participants, with mean scores close to the maximum possible value on the Likert scale. The regression analysis, while explaining a modest portion of the variance (R-squared = 0.126), underscores the critical role of inspirational leadership, supportive supervisors, and a strong sense of belonging in enhancing job satisfaction. These results suggest that organizations aiming to improve employee satisfaction should focus on fostering inspirational leadership, providing robust support from supervisors, and cultivating an inclusive culture where employees feel a strong sense of belonging. This study contributes to the existing literature by highlighting the specific cultural elements that significantly impact employee job satisfaction, providing actionable insights for organizational development and human resource management.

Online updating Huber robust regression for big data streams

ABSTRACT Big data streams have garnered significant attention in multiple industries. However, the immense volume and the presence of outliers in high-velocity streaming data pose great challenges to its analysis. To address these concerns, this paper introduces a novel Online Updating Huber Robust Regression algorithm. By efficiently capturing the salient features of new data subsets, a computationally efficient online updating estimator is proposed without the need for storing historical data. Furthermore, by incorporating Huber regression into its framework, the estimator exhibits robustness to heavy-tailed, heterogeneous as well as outlier-contaminated data. Theoretically, the proposed online updating estimator is asymptotically equivalent to an Oracle estimator derived from the entire dataset. Extensive numerical simulations and a real-world data analysis have been conducted to demonstrate the effectiveness and practicality of the proposed method.

THE DIGITAL ENTERPRISE LANDSCAP: HOW DESI METRICS SHAPE ECONOMIC GROWTH IN THE EU

The European Union’s (EU) economic growth is increasingly driven by digitalization across all areas, including the digital transformation of businesses. This paper explores how the digital transformation of enterprises could influence economic growth in the EU by analyzing data from the Digital Economy and Society Index (DESI) and economic metrics from Eurostat for the post-pandemic period (2021-2022). This paper uses econometric techniques like Panel Corrected Standard Errors (PCSEs) and Robust Regression with Huber Iteration (RRHI) to investigate the cause-and-effect relationships between digitalization and economic factors such as GDP growth rate and real GDP per capita growth. The results show that different digital strategies have varying effects on economic performance. The paper also emphasizes the critical role of real labor productivity in connecting digital transformation with economic performance. The implications of this paper highlight the need for nuanced policy strategies that foster digital skills, improve technological infrastructure and support the Small and Medium Enterprises (SMEs) in their digitalization efforts in order to achieve the ambitious objectives of the EUs Digital Decade. The results also propose a strategic focus on specific technologies that offer immediate economic advantages and enable a stronger incorporation of digital progress across various business operations in the member states.

Evaluating Personalized Add-On Ayurveda Therapy in Oxygen-Dependent Diabetic COVID-19 Patients: A 60-Day Study of Symptoms, Inflammation, and Radiological Changes.

Background Effective management of both acute and post-acute sequelae of SARS-CoV-2 is essential, particularly for type 2 diabetes mellitus (T2DM) patients, who are at increased risk of severe pro-inflammatory responses and complications. Persistent symptoms and residual lung and cardiovascular damage in post-coronavirus disease (COVID-19) individuals highlight the need for comprehensive long-term treatment strategies. Conventional treatments, including Remdesivir and glucocorticoids, have limitations, suggesting that further investigation into Ayurvedic therapies could be beneficial, though controlled trials are currently limited. Objectives Evaluate the effectiveness and safety of Ayurveda with the standard of care (SOC) versus SOC in improving symptoms, moderating immune responses (interleukin-6 (IL-6), C-reactive protein (CRP), neutrophil-lymphocyte ratio (NLR), and radiological outcomes in oxygen-dependent, high-risk, non-vaccinated type 2 diabetes COVID-19 patients over 60 days, and thus addressing their heightened vulnerability to severe infections. Methods A controlled trial with 50 diabetic COVID-19 patients, aged 18-80, with an NLR of >= 4, primarily on Remdesivir, was assigned to Group 1 (Add-on Ayurveda+SOC, n=30) or Group 2 (SOC, n=20) based on their voluntary choice with follow-up on days 14, 28, and 60. Parametric outcomes in group analysis were assessed with robust regression and non-parametric outcomes with Cochran-Mantel-Haenszel, log-rank test, and chi-square tests at 95% confidence interval (CI). Results Group 1 exhibited statistically significant improvements in fever, cough, diarrhea, as well as NLR, IL-6, and CRP by 14 days, and in anosmia, loss of taste, shortness of breath, general weakness, and headache by 60 days. Though the sample size is small, notable improvements can be seen in troponin levels in Group 1 at 28 and 60 days. High-resolution computer tomography COVID-19 reporting and data system (HRCT CO-RADS) scores improved more slowly in Group 2 than in Group 1. Survival rates were 96.4% for Group 1 and 90% for Group 2. Numbers were too small for reliable comparisons at 60 days. Conclusion The add-on Ayurveda group showed a better symptomatic response, and faster normalization in inflammatory markers, including IL-6 and NLR by 14 days, and cardiac markers by 28 days. Minimal clinical and no laboratory adverse events were observed. This study supports the need for a randomized, double-blind trial.

Statistical Reliability Estimation of Earth-Orbiting Satellites Operating from 1991 to 2020

Reliability analysis using satellite failure data for satellites launched in the years 1991–2020 is presented. The analysis is conducted using nonparametric as well as parametric methods. In order to derive a nonparametric reliability estimate from the raw failure data, the Kaplan–Meier estimator is utilized. In order to determine a correct distribution to parameterize the nonparametric estimate, a novel distribution identification framework is applied. The Weibull distribution is consistently identified as best and is then utilized to parameterize the behaviors seen in the nonparametric results. A novel robust regression parameterization method is also applied to improve the traditional regression methodology. Data are split into three decade-long groups with respect to launch dates to investigate time-based trends. Results from this analysis show the most recent decade (2011–2020) to be the worst performing in terms of reliability. This decade in particular is found to be plagued by early or failed reentry cases.

A comparative assessment of Lead (Pb) concentration and physicochemical parameters in groundwater from the Kabwe mine and Lusaka non-mine sites, Zambia

This study addresses the comparative influence of physicochemical parameters on the dissolution of Pb and its levels in the groundwater of former Pb mining areas in Kabwe and non-mining areas in Lusaka, Zambia. This cross-sectional study analyzed 61 groundwater samples from hand pump boreholes using atomic absorption spectrometry. Physicochemical parameters were measured, including electrical conductivity (EC), temperature, total dissolved solids (TDS), and pH. Statistical analysis involved the Wilcoxon test, Spearman’s correlation, robust linear regression, Principal component analysis (PCA), and generalized additive models. Pb concentrations were particularly higher (p = 0.029; median = 0.131 mg/L) in mining areas compared to non-mining areas (median = 0.071 mg/L). The EC in the mining areas (median = 585 µS/cm) was significantly (p = 0.001) lower than those in the non-mining area (median = 1100 µS/cm). Likewise, TDS levels revealed a significant (p = 0.001) difference between the mining area (median = 303 ppm) and the non-mining area (median = 601 ppm). The median pH values for groundwater samples in the mining and non-mining areas were 6.96 and 7.27, respectively. Spearman’s rank correlation revealed a significant (rho = − 0.405; p = 0.036) inverse relationship between Pb concentrations and pH in non-mining areas. Robust linear regression and generalized additive models underscored a significant (p = 0.016) non-linear relationship between temperature and Pb across all areas, with the non-mining area demonstrating a significant (p = 0.003) non-linear relationship between temperature and pH concerning Pb. The study associates mining activities with increased Pb concentrations, underscoring it as a principal contamination source. In contrast, elemental dissolutions from the basement limestone with dolomite might increase Pb concentration in the non-mining areas. The observed positive relationship between temperature and pH in non-mining areas highlights its positive interactions with Pb mobility while temperature greatly influences Pb dissolutions in mining areas.

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.

Depressive symptoms and the processing of unexpected social feedback: Differences in surprise levels, feedback acceptance, and "immunizing" cognition.

Negative social expectations are a key symptom of depression. It has been suggested that individuals with depressive symptoms tend to maintain these expectations by devaluing new experiences that do not fit prior expectations. However, our understanding of the role of such "immunizing" cognition in response to unexpected social feedback in depression, as well as the cognitive mechanisms involved, remains limited. This study investigated the association between depressive symptoms and the cognitive processing of unexpected positive or negative social feedback using a novel, video-based approach featuring naturalistic social stimuli in a subclinical online sample (N = 155). We also examined how surprise levels, feedback acceptance and immunizing cognition relate to other cognitive processes, such as attributional style and rumination, using cross-sectional network analyses. Robust multiple linear regression analyses revealed that depressive symptoms were associated with higher surprise levels (R2adj. = .27), lower feedback acceptance (R2adj. = .19), and higher levels of immunizing cognition (R2adj. = .09) in response to unexpected positive social feedback, but only partially to unexpected negative social feedback. The network analysis suggested that self-efficacy expectations for coping with negative feelings and acceptance of positive social feedback had the strongest expected influence on the different cognitive processes. Our study highlights the challenges that individuals with depressive symptoms face in utilizing positive social feedback to modify negative expectations. For clinicians, our findings suggest the importance of promoting acceptance of positive social feedback, while simultaneously inhibiting immunizing cognition and avoiding the use of overly positive feedback.

Optimizing Electric Vehicle (EV) Charging with Integrated Renewable Energy Sources: A Cloud-Based Forecasting Approach for Eco-Sustainability

As electric vehicles (EVs) are becoming more common and the need for sustainable energy practices is growing, better management of EV charging station loads is a necessity. The simple act of folding renewable power from solar or wind in an EV charging system presents a huge opportunity to make them even greener as well as improve grid resiliency. This paper proposes an innovative EV charging station energy consumption forecasting approach by incorporating integrated renewable energy data. The optimization is achieved through the application of SARLDNet, which enhances predictive accuracy and reduces forecast errors, thereby allowing for more efficient energy allocation and load management in EV charging stations. The technique leverages comprehensive solar and wind energy statistics alongside detailed EV charging station utilization data collected over 3.5 years from various locations across California. To ensure data integrity, missing data were meticulously addressed, and data quality was enhanced. The Boruta approach was employed for feature selection, identifying critical predictors, and improving the dataset through feature engineering to elucidate energy consumption trends. Empirical mode decomposition (EMD) signal decomposition extracts intrinsic mode functions, revealing temporal patterns and significantly boosting forecasting accuracy. This study introduces a novel stem-auxiliary-reduction-LSTM-dense network (SARLDNet) architecture tailored for robust regression analysis. This architecture combines regularization, dense output layers, LSTM-based temporal context learning, dimensionality reduction, and early feature extraction to mitigate overfitting. The performance of SARLDNet is benchmarked against established models including LSTM, XGBoost, and ARIMA, demonstrating superior accuracy with a mean absolute percentage error (MAPE) of 7.2%, Root Mean Square Error (RMSE) of 22.3 kWh, and R2 Score of 0.87. This validation of SARLDNet’s potential for real-world applications, with its enhanced predictive accuracy and reduced error rates across various EV charging stations, is a reason for optimism in the field of renewable energy and EV infrastructure planning. This study also emphasizes the role of cloud infrastructure in enabling real-time forecasting and decision support. By facilitating scalable and efficient data processing, the insights generated support informed energy management and infrastructure planning decisions under dynamic conditions, empowering the audience to adopt sustainable energy practices.

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.

Weighted penalized m-estimators in robust ridge regression: an application to gasoline consumption data

ABSTRACT The OLS and ridge regression (RR) estimators are adversely affected, when the problem of multicollinearity and y-direction outliers occur together. The robust ridge regression with penalized parameters offers biased estimators with lower variance than OLS and RR. The optimal penalized parameter value is crucial for balancing variance and bias. This study proposes three new weighted robust penalized M-estimators to address both multicollinearity and y-direction outliers. Their performance is evaluated against OLS, RR and existing penalized M-estimators using Monte Carlo simulations based on mean squared error (MSE). The new estimators exhibit lower MSE in the presence of multicollinearity and y-direction outliers. A real data application on gasoline consumption demonstrates the superior performance of the new estimators.