Variance Inflation Factor Research Articles

Landslide susceptibility assessment along the Karakoram highway, Gilgit Baltistan, Pakistan: A comparative study between ensemble and neighbor-based machine learning algorithms

This study addressed the complex challenges associated with landslide detection along the Karakoram Highway (KKH), where tectonic events and data availability limitations posed significant obstacles. To overcome these hurdles, the research framework encompassed several critical components. First, it tackled the issue of multicollinearity through the application of statistical measures such as Variable Inflation Factor (VIF) and Information Gain (IG). Secondly, the study emphasized the importance of selecting a study area that would comprehensively represent the multivariate landscape, with KKH serving as an illustrative example. In striving for an equilibrium between implementation ease and algorithmic performance, the research favored the adoption of Random Forest (RF) and Extremely Randomized Trees (EXT) over XGBoost. Lastly, to fine-tune the algorithms and optimize their parameters, the study employed Particle Swarm Optimization (PSO) and evaluated their performance using metrics like the Area Under the Curve (AUC). Remarkably, this comprehensive approach yielded accuracy rates exceeding 90% for all algorithms tested (RF, EXT, and K-Nearest Neighbor (KNN)), with specific AUC values of 0.967, 0.968, and 0.914, respectively. These findings offer invaluable insights into enhancing disaster prevention strategies and informing land-use planning efforts along the KKH highway.

Nomogram establishment for short-term survival prediction in ICU patients with aplastic anemia based on the MIMIC-IV database

ABSTRACT Objective To establish an efficient nomogram model to predict short-term survival in ICU patients with aplastic anemia (AA). Methods The data of AA patients in the MIMIC-IV database were obtained and randomly assigned to the training set and testing set in a ratio of 7:3. Independent prognosis factors were identified through univariate and multivariate Cox regression analyses. The variance inflation factor was calculated to detect the correlation between variables. A nomogram model was built based on independent prognostic factors and risk scores for factors were generated. Model performance was tested using C-index, receiver operating characteristic (ROC) curve, calibration curve, decision curve analysis (DCA) and Kaplan-Meier curve. Results A total of 1,963 AA patients were included. A nomogram model with 7 variables was built, including SAPS II, chronic pulmonary obstructive disease, body temperature, red cell distribution width, saturation of peripheral oxygen, age and mechanical ventilation. The C-indexes in the training set and testing set were 0.642 and 0.643 respectively, indicating certain accuracy of the model. ROC curve showed favorable classification performance of nomogram. The calibration curve reflected that its probabilistic prediction was reliable. DCA revealed good clinical practicability of the model. Moreover, the Kaplan-Meier curve showed that receiving mechanical ventilation could improve the survival status of AA patients in the short term but did not in the later period. Conclusion The nomogram model of the short-term survival rate of AA patients was built based on clinical characteristics, and early mechanical ventilation could help improve the short-term survival rate of patients.

Predictive model for non-malignant portal vein thrombosis associated with cirrhosis based on inflammatory biomarkers.

Portal vein thrombosis (PVT), a complication of liver cirrhosis, is a major public health concern. PVT prediction is the most effective method for PVT diagnosis and treatment. To develop and validate a nomogram and network calculator based on clinical indicators to predict PVT in patients with cirrhosis. Patients with cirrhosis hospitalized between January 2016 and December 2021 at the First Hospital of Lanzhou University were screened and 643 patients with cirrhosis who met the eligibility criteria were retrieved. Following a 1:1 propensity score matching 572 patients with cirrhosis were screened, and relevant clinical data were collected. PVT risk factors were identified using the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression analysis. Variance inflation factors and correlation matrix plots were used to analyze multicollinearity among the variables. A nomogram was constructed to predict the probability of PVT based on independent risk factors for PVT, and its predictive performance was verified using a receiver operating characteristic curve (ROC), calibration curves, and decision curve analysis (DCA). Finally, a network calculator was constructed based on the nomograms. This study enrolled 286 cirrhosis patients with PVT and 286 without PVT. LASSO analysis revealed 13 variables as strongly associated with PVT occurrence. Multivariate logistic regression analysis revealed nine indicators as independent PVT risk factors, including etiology, ascites, gastroesophageal varices, platelet count, D-dimer, portal vein diameter, portal vein velocity, aspartate transaminase to neutrophil ratio index, and platelet-to-lymphocyte ratio. LASSO and correlation matrix plot results revealed no significant multicollinearity or correlation among the variables. A nomogram was constructed based on the screened independent risk factors. The nomogram had excellent predictive performance, with an area under the ROC curve of 0.821 and 0.829 in the training and testing groups, respectively. Calibration curves and DCA revealed its good clinical performance. Finally, the optimal cutoff value for the total nomogram score was 0.513. The sensitivity and specificity of the optimal cutoff values were 0.822 and 0.706, respectively. A nomogram for predicting PVT occurrence was successfully developed and validated, and a network calculator was constructed. This can enable clinicians to rapidly and easily identify high PVT risk groups.

Logistic regression analysis of risk factors for pediatric burns: a case-control study in underdeveloped minority areas in China.

Pediatric burns are common, especially in underdeveloped countries, and these can physically affect the children involved and have an impact on their mental health. The aim of the present study was to assess the effect of pediatric burns in underdeveloped minority areas of China. Case information from 192 children was collected from outpatient and inpatient clinics using a survey questionnaire. These included 90 pediatric burn cases and 102 controls who were children without burns. A stepwise logistic regression analysis was used to determine the risk factors for pediatric burns in order to establish a model. The goodness-of-fit for the model was assessed using the Hosmer and Lemeshow test as well as receiver operating characteristic and internal calibration curves. A nomogram was then used to analyze the contribution of each influencing factor to the pediatric burns model. Seven variables, including gender, age, ethnic minority, the household register, mother's employment status, mother's education and number of children, were analyzed for both groups of children. Of these, age, ethnic minority, mother's employment status and number of children in a household were found to be related to the occurrence of pediatric burns using univariate logistic regression analysis (p < 0.05). After a collinearity diagnosis, a multivariate logistic regression analysis of variables with tolerances of >0.2 and variance inflation factor <5 showed that age was a protective factor for pediatric burns [odds ratio (OR) = 0.725; 95% confidence interval (CI): 0.665-0.801]. Compared with single-child parents, those with two children were at greater risk of pediatric burns (OR = 0.389; 95% CI: 0.158-0.959). The ethnic minority of the child and the mother's employment status were also risk factors (OR = 6.793; 95% CI: 2.203-20.946 and OR = 2.266; 95% CI: 1.025-5.012, respectively). Evaluation of the model used was found to be stable. A nomogram showed that the contribution in the children burns model was age > mother's employment status > number of children > ethnic minority. This study showed that there are several risk factors strongly correlated to pediatric burns, including age, ethnic minority, the number of children in a household and mother's employment status. Government officials should direct their preventive approach to tackling the problem of pediatric burns by promoting awareness of these findings.

A framework for optimizing environmental covariates to support model interpretability in digital soil mapping

A common practice in digital soil mapping (DSM) is to incorporate many environmental covariates into a machine-learning algorithm to predict the spatial patterns of soil attributes. Variance inflation factor (VIF), principal component analysis (PCA), and recursive feature elimination (RFE) are three statistical methods that can be used to reduce the number of covariates. This study aims 1) to compare VIF and PCA approaches; 2) to identify an approach to determine the minimum number of covariates in DSM to ensure model parsimony using RFE after using VIF; and 3) to examine methods to interpret the impact of covariates on the variability of the predicted soil properties. The study area was the province of British Columbia (BC), Canada. This study used legacy data for four soil properties to make digital soil maps: soil organic carbon (SOC%), pH, clay%, and coarse fragment (CF%). Seven models were made for each soil property to determine the influence on validation results by using a different number of covariates produced by various methods on validation results. The results showed that the number of covariates could be reduced from 70 to 4 to 12 with only a little or no difference in concordance correlation coefficient (CCC) validation results. The CCC results of pH models using 70 and 7 covariates were both 0.74, and for other soil properties, this difference was negligible. The validation results obtained from PCA models showed that the performance of PCA in reducing the number of covariates was not as effective as when using VIF. Moreover, this study showed that covariates related to precipitation were the most important for modeling SOC%, soil pH, and clay%. Topographic covariates were the most influential covariates for modeling soil CF%. This study emphasizes the potential benefits of combining various data reduction methods to achieve optimal outcomes and generate the most parsimonious and interpretable models.

Family Satisfaction with Care Provided in Intensive Care Unit; a Multi-Center, Cross-Sectional Study.

Healthcare provided in medical facilities should prioritize the needs of families, as it enhances the quality of care for the patients. Family satisfaction gauges how effectively healthcare professionals address the perceived needs and expectations of family members. Numerous factors, including information dissemination, communication, family dynamics, patient characteristics, hospital facilities, and the caregiving process, serve as predictors of family satisfaction. Thus, this study seeks to evaluate the satisfaction of families with the care received by patients admitted to the intensive care unit. A hospital-based cross-sectional study involving 400 participants was conducted across multiple centers from March to June 2023. Multicollinearity was assessed by examining variance inflation factors (VIF), while the goodness-of-fit was evaluated using the Hosmer and Lemeshow test. Both bivariable and multivariable logistic regression analyses were utilized to identify factors correlated with family satisfaction. Variables with a p-value below 0.2 in the bivariable logistic regression were included in the multivariable logistic regression analysis. Adjusted Odds Ratios (AORs) with 95% Confidence Intervals were computed to indicate the strength of association. In the multivariable analysis, variables with a p-value less than 0.05 were deemed statistically significant. The overall family satisfaction with the care provided in the intensive care unit was 58.6%, with a 95% confidence interval ranging from 55.882% to 61.241%. Families expressed higher satisfaction levels with patient care (64.8%) and professional care (67.4%). However, they reported lower satisfaction levels regarding care provided for families (52.2%), the ICU environment (56.8%), and involvement of families in decision-making (55.8%). Lack of formal education (AOR: 1.949, 95% CI: 1.005, 4.169), completion of primary education (AOR: 2.581, 95% CI: 1.327, 5.021), and completion of grades 9-12 (AOR: 2.644, 95% CI: 1.411, 4.952) were found to be significantly associated with overall family satisfaction. The overall level of satisfaction is satisfactory. To enhance service quality and family satisfaction, healthcare providers should prioritize effective and regular communication with family members. Keeping them well informed about the patient's condition and treatment plan is essential.

Modeling of Gamma Index for Prediction of Pretreatment Quality Assurance in Stereotactic Body Radiation Therapy of the Liver.

The purpose of this study was to develop a predictive model to evaluate pretreatment patient-specific quality assurance (QA) based on treatment planning parameters for stereotactic body radiation therapy (SBRT) for liver carcinoma. We retrospectively selected 180 cases of liver SBRT treated using the volumetric modulated arc therapy technique. Numerous parameters defining the plan complexity were calculated from the DICOM-RP (Radiotherapy Plan) file using an in-house program developed in MATLAB. Patient-specific QA was performed with global gamma evaluation criteria of 2%/2 mm and 3%/3 mm in a relative mode using the Octavius two-dimensional detector array. Various statistical tests and multivariate predictive models were evaluated. The leaf speed (MILS) and planning target volume size showed the highest correlation with the gamma criteria of 2%/2 mm and 3%/3 mm (P < 0.05). Degree of modulation (DoM), MCSSPORT, leaf speed (MILS), and gantry speed (MIGS) were predictors of global gamma pass rate (GPR) for 2%/2 mm (G22), whereas DoM, MCSSPORT, leaf speed (MILS) and robust decision making were predictors of the global GPR criterion of 3%/3 mm (G33). The variance inflation factor values of all predictors were <2, indicating that the data were not associated with each other. For the G22 prediction, the sensitivity and specificity of the model were 75.0% and 75.0%, respectively, whereas, for G33 prediction, the sensitivity and specificity of the model were 74.9% and 85.7%%, respectively. The model was potentially beneficial as an easy alternative to pretreatment QA in predicting the uncertainty in plan deliverability at the planning stage and could help reduce resources in busy clinics.

A Redefined Variance Inflation Factor: Overcoming the Limitations of the Variance Inflation Factor

A Redefined Variance Inflation Factor: Overcoming the Limitations of the Variance Inflation Factor

Tax Planning, Corporate Tax Avoidance and Real Earnings Management: GMM Approach in Nigeria

The study examines the effect of tax planning and corporate tax avoidance on Real Earnings Management (REM) in Nigeria. Data were gathered from the financial reports of 58 listed non-financial firms for twelve years, from 2010 to 2021. Preliminary analyses such as descriptive statistics, correlation matrix, and variance inflation factor were conducted. The Generalized Method of Moments (GMM) was employed to identify a possible effect on REM. The study revealed that tax planning has a positive and significant effect on REM, while corporate tax avoidance has a negative and significant influence on REM. Also, leverage revealed a negative significant effect on REM, while firm growth has no significant effect on REM. However, firm size has a significant positive influence on REM. The study concludes that tax planning and corporate tax avoidance have a significant influence on REM. Thus, the study recommends that tax authorities should ensure that firm management engages in lowering capitalization limits, ensures first-in, first-out instead of last-in, reduces nonobligatory expenses for short periods, and ensures moderate deferred tax expenses in order to reduce REM through tax planning.

Impact of Personality Traits and Sustainability Orientation on Social Entrepreneurial Intentions among Engineering Graduates: A Test of the Big Five Personality Approach

The study intends to explore the relationship that exists between the big five personality traitsagreeableness, conscientiousness, extraversion, neuroticism, openness, and sustainability orientationand the intention of engineering students to engage in social entrepreneurship. The data were analyzed using SMART PLS software. A five-point Likert scale questionnaire was distributed to 259 samples both in-person and online. Structural equation modelling was used to examine the impact of the big five personality traits and sustainability orientation on the intention to engage in social entrepreneurship.The reliability was assessed using Cronbach Alpha and Composite Reliability (CR), the multicollinearity was assessed using the Variance Inflation Factor (VIF), and the discriminant validity was assessed using the Fornell and Larcker criterion and the Hetero Trait-Mono Trait (HTMT) ratio. The study identified that social entrepreneurship intention is positively and significantly impacted by the proxies of the big five personality traits, including openness, agreeableness, extraversion, and sustainability orientation. The research indicates that graduates from engineering whohavea strong focus on sustainability orientation, alongside their personality traits, tend to demonstrate an elevated desire and intention to set up social enterprises. The findings of the study provide policymakers valuable insights into crafting targeted incentives for graduates, aimed at fostering the growth of social enterprises.

Application of nonparametric quantifiers for online handwritten signature verification: A statistical learning approach

AbstractThis work explores the use of nonparametric quantifiers in the signature verification problem of handwritten signatures. We used the MCYT‐100 (MCYT Fingerprint subcorpus) database, widely used in signature verification problems. The discrete‐time sequence positions in the x ‐axis and y‐axis provided in the database are preprocessed, and time causal information based on nonparametric quantifiers such as entropy, complexity, Fisher information, and trend are employed. The study also proposes to evaluate these quantifiers with the time series obtained, applying the first and second derivatives of each sequence position to evaluate the dynamic behavior by looking at their velocity and acceleration regimes, respectively. The signatures in the MCYT‐100 database are classified via Logistic Regression, Support Vector Machines (SVM), Random Forest, and Extreme Gradient Boosting (XGBoost). The quantifiers were used as input features to train the classifiers. To assess the ability and impact of nonparametric quantifiers to distinguish forgery and genuine signatures, we used variable selection criteria, such as: information gain, analysis of variance, and variance inflation factor. The performance of classifiers was evaluated by measures of classification error such as specificity and area under the curve. The results show that the SVM and XGBoost classifiers present the best performance.

Forest fire vulnerability in Nepal's Chure region: Investigating the influencing factors using generalized linear model

The Chure region, among the world's youngest mountains, stands out as highly susceptible to natural calamities, particularly forest fires. The region has consistently experienced forest fire incidents, resulting in the degradation of valuable natural and anthropogenic resources. Despite its vulnerability, there have been limited studies to understand the relationship of various causative factors for the recurring fire problem. Hence, to comprehend the influencing factors for the recurring forest fire problem and its extent, we utilized generalized linear modeling under binary logistic regression to combine the dependent variable of satellite detected fire points and various independent variables. We conducted a variance inflation factor (VIF) test and correlation matrix to identify the 14 suitable variables for the study. The analysis revealed that forest fires occurred mostly during the three pre-monsoon periods and had a significant positive relation with the area under forest, rangeland, bare-grounds, and Normalized Difference Vegetation Index (NDVI) (P < 0.05). Consequently, our model showed that the probability of fire incidents decreases with elevation, precipitation, and population density (P < 0.05). Among the significant variables, the forest areas emerges as the most influencing factor, followed by precipitation, elevation, area of rangeland, population density, NDVI, and the area of bare ground. The validation of the model was done through the area under the curve (AUC = 0.92) and accuracy (ACC = 0.89) assessments, which showed the model performed excellently in terms of predictive capabilities. The modeling result and the forest fire susceptible map provide valuable insights into the forest fire vulnerability in the region, offering baseline information about forest fires that will be helpful for line agencies to prepare management strategies to further prevent the deterioration of the region.

EXPERIMENTAL RELATIONSHIP BETWEEN LAND USE/LAND COVER AND UNDERGROUND FLOW IN AL BA COMMUNE IN THE CENTRAL REGION OF VIETNAM

Rapid urbanization and intensive agricultural practices in Vietnam's Central Highlands have substantially transformed land use/land cover (LULC) patterns, with implications for groundwater resources necessitating measurement. The objective of this investigation is to examine the links between multi-temporal variations in LULC and changes in underground water flow in Al Ba Commune, Gia Lai Province. Supervised object-based classification of Landsat and Sentinel imagery from the past 25 years has produced consistent LULC maps as inputs for the model. Geospatial data of LULC, meteorological data, and in-situ data of groundwater flow (Q) were used to calibrate and validate the integrated hydrological MIKE SHE model and simulate past Q conditions for correlation analysis. With these data on LULC and groundwater flow, this study investigates the relationship between LULC changes and Q using Multivariate Linear Regression (MLR) and Ridge Regression models. The MLR model showed high R² (0.91) but positive impacts of all LULC classes on Q, contrary to expectations. Further analysis with Variance Inflation Factor (VIF) revealed multicollinearity among LULC variables, which was addressed using Ridge Regression with a bias factor to correct for this issue. The revised model demonstrated that non-vegetative covers negatively affect Q while vegetative covers have a positive impact. These results highlight the complex dynamics of LULC changes on groundwater resources, which can be detected by statistical approaches. The results of study could be important for sustainable land management.

Estimation of wheat protein content and wet gluten content based on fusion of hyperspectral and RGB sensors using machine learning algorithms

The protein content (PC) and wet gluten content (WGC) are crucial indicators determining the quality of wheat, playing a pivotal role in evaluating processing and baking performance. Original reflectance (OR), wavelet feature (WF), and color index (CI) were extracted from hyperspectral and RGB sensors. Combining Pearson-competitive adaptive reweighted sampling (CARs)-variance inflation factor (VIF) with four machine learning (ML) algorithms were used to model accuracy of PC and WGC. As a result, three CIs, six ORs, and twelve WFs were selected for PC and WGC datasets. For single-modal data, the back–propagation neural network exhibited superior accuracy, with estimation accuracies (WF > OR > CI). For multi-modal data, the random forest regression paired with OR + WF + CI showed the highest validation accuracy. Utilizing the Gini impurity, WF outweighed OR and CI in the PC and WGC models. The amalgamation of MLs with multimodal data harnessed the synergies among various remote sensing sources, substantially augmenting model precision and stability.

Abstract P493: A Multivariate Analysis of Genetic Risk Scores for Heart Failure and Left Ventricular Mass Derived From Multi-Ancestry Populations and Left Ventricular Mass Amongst African Americans: The Jackson Heart Study

Background: Increased left ventricular (LV) mass is an outcome of LV remodeling that can result in development of left ventricular hypertrophy (LVH), a maladaptive disorder that increases the risk of heart failure. Elevated LV mass is a sensitive predictor of adverse cardiovascular events. African Americans experience a disproportionate burden of heart failure risk and development of LVH. The influence of genetics on this health disparity is unclear and a topic of interest. Genome-wide association studies (GWAS) have uncovered many variants associated with numerous adverse cardiovascular outcomes, such as heart failure and LV phenotypes. However, the relevance of these studies for African Americans may be limited as these analyses have been conducted in multi-ancestry populations with limited representation of African heritage. The question of whether these genetic risk variants can be used to predict LV phenotypes in an independent African American population remains unanswered. Hypothesis: Variant loci that have been associated with heart failure and LV mass in multi-ancestry populations will be associated with LV mass amongst African Americans in the Jackson Heart Study (JHS). Methods: Thirty-nine (39) SNPs associated with heart failure and ten (10) SNPs associated with LV mass were identified from two published studies in which participants of African ancestry or Black individuals comprised less than 15% and 2%, respectively. Using the published effect size estimates, genetic risk scores (GRSs) for heart failure and LV mass were computed for 2,175 African American participants (mean age 55, 63% female) in the JHS with complete data and consent for genetic research. We assessed the linear association of heart failure and LV mass GRSs with LV mass indexed to height (LVM h ) and indexed to body surface area (LVM bsa ). A multivariate general linear model was fit to both LVM h and LVM bsa containing both GRS predictors and accounting for age, sex, BMI (mean 31.5), diabetes (22%), and hypertension (52%). No multicollinearity was noted among these explanatory variables using variance inflation factors. Due to non-constant error variance, a log transformation was conducted on both LV mass response variables. Type III MANOVA Pillai tests were run to assess the effects of the GRSs on the log LVM bsa and log LVM h values. Conclusion: When accounting for age, BMI, sex, diabetes, and hypertension, we found that neither the heart failure GRS (p-val = 0.4012) nor the LV mass GRS (p-val = 0.8332) had an effect on both log LVM bsa and log LVM h . There is insufficient evidence to suggest that heart failure and LV mass variants derived from predominantly European multi-ancestry populations are linearly associated with log LV mass amongst African Americans in the Jackson Heart Study. Our work highlights the need for greater recruitment and targeted studies of populations with African ancestry for GWAS of LV phenotypes.

Assessment of Small Scale Mine Blast Toe Volume Effect on Fragmentation Size Distribution: An Application of Edge Detection Software

This study explores the critical aspect of blast toe volume in small-scale mining operations and its impact on fragmentation size distribution. The assessment is conducted to understand the relationship between blast toe volume and the resultant blast design patterns. A comprehensive analysis is carried out using production blast result from four dolomite quarries in Akoko Edo, Nigeria, focusing on various explosive engineering parameters. The research employs advanced measurement techniques and statistical methods to quantify blast toe volume and assess its influence on fragmentation size distribution. By systematically varying blast toe volumes in controlled experiments, the study aims to establish correlations between toe volume and the resulting fragmentation size. The Variance inflation factor obtained in this study revealed that the selection of parameters during toe volume simulation must be carried out with respect to stemming length and explosive weight (MIC). The result shows that the maximum instantaneous charge has a negative correlation influence on toe volume as stemming also increases. This reveals that variation in stemming length results in low explosive energy release along the blast hole column, causing toe undulation. Finally, it was also revealed that at the blast mean size (X50) increases, the toe also increases due to the poor utilization of explosive energy at the blast column.

Prediction of non-uniform shrinkage of steel-concrete composite slabs based on explainable ensemble machine learning model

The non-uniform shrinkage (NUS) of steel-concrete composite slabs has been proved to significantly affect the long-term performance of composite slabs. But the reliable analysis method has not yet been proposed due to the multi-factor coupling mechanism of material composition and boundary conditions. A database containing 782 samples was established in this paper, and the index set containing 5 feature parameters was selected by analyzing the physical mechanism of NUS of steel-concrete composite slabs. The predictive capabilities of 8 machine learning (ML) models with different working mechanisms were compared, and the reliability of ML models was verified by Taylor and box scatter from the perspective of global and single sample, respectively. Furthermore, sensitivity analysis was conducted using the SHapley Additive exPlanations (SHAP) method to evaluate the impact of each parameter on the NUS of steel-concrete composite slabs. Finally, a user-friendly graphical user interface (GUI) was developed to guide engineering design. The limitations and potential improvements of the ML models established in this paper were discussed. The results indicate that the variance inflation factor (VIF) method and correlation analysis can effectively mitigate the influence of data multicollinearity and variable collinearity, and the pearson coefficient and VIF value of the input variables are less than 0.35 and 5. Gradient boosting decision tree (GBDT) has the best prediction performance, with R2 of 0.933 and 0.927 on the training set and testing set, respectively. Relative distance (x) and age (t) are the key parameters affecting the NUS of steel-concrete composite slabs. The procedure and GUI developed in this paper can effectively guide the intelligent prediction of the composite slab NUS.

Optimization and Impact Assessment of Technology Management of Solar Energy in Rwanda

There is a rarity of research about technology management in the solar energy industry in Rwanda and very little analysis is available on the effectiveness of the technology used in solar home systems (SHSs). Using a binary logistic regression to primary data collected through Kobo Collect, this study assesses the factors that are involved in SHS business in Rwanda. The findings reveal a difference in successfully collecting money from customers who purchase SHSs on an installment payment plan concerning the type of SHS (i.e. pay-as-you-go (PAYG) and non-pay-as-you-go (non-PAYG)). In terms of the odds ratio concept, a statistical interpretation was provided. If the customer has a non-PAYG SHS, getting a reminder from the solar energy company’s agent will increase by 9.52 times. That means, it is 9.52 times higher for the non-PAYG SHS as compared to a customer with PAYG SHS. The results also show that the multicollinearity assumptions for the economic, social, technological, and management models are not violated and are optimal because each of the variance inflation factors (VIFs) was close to unity. However, only the socialization predictors were not significant enhancements in the goodness-of-fit relative to the intercept-only model. The Nagelkerke R2 indicates only an average (49.5%) relationship between the prediction and grouping of the modified economic pricing model. The Hosmer–Lemeshow measures indicate that the models in this research are optimal and good fits for the data studied. Furthermore, the predicted success overall performance for the economic, social, technological, and managerial models was 73.2% on average. The Wald criterion equally demonstrated that only the monthly installment amount (with and without outliers), type of SHS, and payment channel made significant contributions to the prediction optimization problem. Therefore, the solar energy companies in Rwanda need to rely on the findings of studies like this to successfully manage customers’ accounts optimally.

Leveraging Remotely Sensed and Climatic Data for Improved Crop Yield Prediction in the Chi Basin, Thailand

Predictions of crop production in the Chi basin are of major importance for decision support tools in countries such as Thailand, which aims to increase domestic income and global food security by implementing the appropriate policies. This research aims to establish a predictive model for predicting crop production for an internal crop growth season prior to harvest at the province scale for fourteen provinces in Thailand’s Chi basin between 2011 and 2019. We provide approaches for reducing redundant variables and multicollinearity in remotely sensed (RS) and meteorological data to avoid overfitting models using correlation analysis (CA) and the variance inflation factor (VIF). The temperature condition index (TCI), the normalized difference vegetation index (NDVI), land surface temperature (LSTnighttime), and mean temperature (Tmean) were the resulting variables in the prediction model with a p-value &lt; 0.05 and a VIF &lt; 5. The baseline data (2011–2017: June to November) were used to train four regression models, which revealed that eXtreme Gradient Boosting (XGBoost), random forest (RF), and XGBoost achieved R2 values of 0.95, 0.94, and 0.93, respectively. In addition, the testing dataset (2018–2019) displayed a minimum root-mean-square error (RMSE) of 0.18 ton/ha for the optimal solution by integrating variables and applying the XGBoost model. Accordingly, it is estimated that between 2020 and 2022, the total crop production in the Chi basin region will be 7.88, 7.64, and 7.72 million tons, respectively. The results demonstrated that the proposed model is proficient at greatly improving crop yield prediction accuracy when compared to a conventional regression method and that it may be deployed in different regions to assist farmers and policymakers in making more informed decisions about agricultural practices and resource allocation.

Understanding Customer Satisfaction Factors: A Logistic Regression Analysis

Background: In the dynamic landscape of modern business, understanding customer satisfaction is crucial for success and competitiveness. This research intends to analyze the various elements that affect customer contentment, concentrating on the impact of demographic characteristics, economic inclinations, and involvement in loyalty programs. Methods: Employing logistic regression analysis, this research analyzes data collected from a diverse customer base across various industries. The study explores the relationship between customer satisfaction (binary dependent variable) and key independent variables, including age, income level, and loyalty program participation. Model validation is conducted through the Hosmer-Lemeshow test, along with the assessment of model fit using Cox &amp; Snell and Nagelkerke R² metrics. Multicollinearity is checked using the Variance Inflation Factor (VIF). Results: The logistic regression model reveals that age and income level significantly influence customer satisfaction, with younger customers and individuals with greater income levels being more probable to report satisfaction. Additionally, participation in loyalty programs emerges as a strong predictor of customer satisfaction. The model demonstrates good fit and predictive ability, as indicated by statistical tests and graphical analyses, including an ROC curve and a Predicted Probability Plot. Conclusions: The research offers significant understanding into the determinants of customer satisfaction, highlighting the importance of demographic factors, economic status, and loyalty programs. These findings offer both theoretical contributions to the field of customer satisfaction research and practical implications for business strategies focused on customer engagement and loyalty.&#x0D; &#x0D; Received: 17 November 2023 / Accepted: 17 February 2024 / Published: 5 March 2024