Search

Classification as an essential part of Machine Learning and Data Mining has significant roles in engineering, medicine, agriculture, military, etc. With the evolution of data collection tools and the unceasing efforts of researchers, new datasets with huge dimensions are obtained so that each data sample has multiple labels. This kind of classification is called Multi-Class Classification (MLC) and demands new techniques to predict the set of labels for a data instance. To date, a variety of methods have been proposed to solve MLC problems. However, new high-dimensional datasets with challenging patterns are being developed, making it necessary for new research to be conducted to develop more efficient methods. This paper presents a novel framework named QLHA to solve MLC problems more efficiently. In the QLHA, the Principal Label Space Transformation (PLST) and Ridge Regression (RR) are recruited to predict the labels of data. Next, an effective objective function is introduced. Also, a hybrid metaheuristic algorithm called QGTOJS is provided to optimize objective value and enhance the predicted labels by selecting the most relevant features. In the QGTOJS, the Gorilla Troops Optimization (GTO) and Jellyfish Search algorithm (JS) are binarized and hybridized through a modified variant of the Q-learning algorithm. Besides, an adjusted Hill Climbing strategy is adopted to balance the exploration and exploitation and improve local optima departure. Likewise, a local search mechanism is provided to enhance searchability as much as possible. Eventually, the QLHA is applied to ten multi-label datasets and the obtained results are compared with heuristic and metaheuristic-based MLC methods numerically and visually. The experimental results disclosed the effectiveness of the contributions and superiority of the QLHA over competitors. [ABSTRACT FROM AUTHOR]

The primary objective of the paper is to predict the price of the "Toyota Cars" based on the requirement of the customer. The primary need is to identify the minimum price of the "Toyota Cars" based on the 37 key attributes/factors. The levels of each factor in the data set varies from 2 to 312, moreover the data sets used for the analysis, contain both feature as well as numerical based data. The data sets are taken from the "KAGGLE" an open-source real time data provider. The total number of data available for the analysis is 1442, since the data are large and the factors are more with different levels, the prediction done using the Machine Learning (ML) approach. Ridge regression, a supervised machine learning algorithm is used to predict the sales price of the "Toyota Cars", thereby identifying the minimum price based on different combinations of attributes. The prime novelty lies on the usage of the Ridge regression, a linear regression algorithm for the prediction process. The ridge regression model was developed by means of the Python programming and Google Colab, an open-source online compiler for the ML was used for the code compilation process. Based on the result, the minimum price was found to be 6520.8 dollars, with the basic customer requirements such as Model: Corolla 1.6-16v, Type: Sedan, 4 Doors, CNG, 69hp, 1854 cc, with air conditioner, centre lock and power steering. [ABSTRACT FROM AUTHOR]

Air pollution is defined as an adverse event that negatively affects ecosystems and standard conditions necessary for human survival and progress, manifested by certain substances in the atmosphere exceeding specific concentration levels. The control of air pollution is a significant strategic task related to the national economies and the well-being of the people. In the face of increasingly severe environmental pollution problems, accurately predicting air pollution indicators becomes crucial. Among the popular air pollution prediction methods, the K-nearest neighbors (KNN) appears to be one of most promising approaches. In this paper, we develop a novel KNN rule that combines the ridge estimators called KNN-ridge regression (KNN-RR). The proposed KNN-RR is motivated by the sensitivity problem that multi-collinearity exists in the current KNN regression, aiming to enhance the prediction performance. Our theoretical result shows that under some mild assumptions, there exists a penalty parameter such that the mean square prediction error of ridge regression is smaller than that of ordinary least square regression. We examine the empirical performances of KNN-RR and other methods on real-world datasets, such as the AQI and PM2.5 prediction, and the results indicate that our method has some advantages in improving prediction accuracy. To a certain extent, this paper paves a new way to improve some supervised machine learning methods. [ABSTRACT FROM AUTHOR]

The global shift towards electric vehicles (EVs) underscores the critical need for reliable battery performance and safety. Lithium-ion batteries, particularly Li-NMC (lithium nickel manganese cobalt oxide), are widely adopted for their balanced functional and performance characteristics. However, the advancement of batteries with higher nickel content and reduced manganese and cobalt introduces challenges, including increased susceptibility to thermal runaway and degradation, especially under abusive conditions like over-discharge. This study addresses significant research gaps by developing a machine learning (ML) algorithm for the early detection and predictive maintenance of over-discharged Li-NMC batteries. Current methods often fail to identify and mitigate the effects of continuous cycling, which can release harmful free radicals such as singlet oxygen (1 O 2 ) and superoxide ( O 2 - ) that accelerate degradation. Our ML approach utilizes supervised learning, feature engineering, and model optimization, leveraging key input features such as voltage, time, and cycle count which are derived from extensive battery life testing. To validate our model, we conducted scanning electron microscopy energy-dispersive spectroscopy (SEM–EDS), galvanostatic charge–discharge (GCD) tests, and rate capability tests. The proposed ridge regression model achieved a mean absolute error (MAE) of 0.11422%, a mean squared error (MSE) of 0.02313%, and an R-squared (R2) value of 0.99, outperforming other models such as Decision Trees (DT), Recurrent Neural Networks (RNNs), Support Vector Machines (SVMs), Gradient Boosting (GB), and Lasso Regression. Our model addresses key shortcomings of existing methods, particularly in predicting degradation in precycled batteries subjected to fault induction. The insights gained contribute to a robust control strategy for EV battery management, enabling proactive maintenance, timely battery replacement, and enhanced system reliability and safety, effectively addressing the long-term challenges in battery health management. [ABSTRACT FROM AUTHOR]

In contrast to multi-class classification, in multi-label classification (MLC), each data sample is a member of several classes. In this regard, the classifiers should discover more complex patterns within the data; accordingly, the complexity grows remarkably. Recently, scholars used the binary relevance, the label powerset, and the ensemble method (EM) techniques to extend state-of-the-art classifiers or develop new ones to solve real-world MLC problems more optimally. The current paper recruits the ridge regression (RR) and a decision tree (DT) and presents a new MLC model named PHMLC. In the PHMLC, the Laplacian Eigenmaps Nonlinear Dimensionality Reduction Algorithm is employed to omit the redundant and irrelevant features. Next, the data is encoded by the principal label space transformation (PLST) and singular value decomposition. Then, the RR is recalled to predict the first set of labels through the encoded data. Besides, the Binary Relevance technique is applied to the original DT, and an MLC model called BR-DT is introduced. The BR-DT is responsible for predicting the second set of labels. Eventually, the greedy selection mechanism is utilized to obtain final labels based on the predicted labels by the PLST-based RR and the BR-DT models. To demonstrate the PHMLC's effectiveness, the PHMLC model is applied to nine real-life multi-label datasets, and the results are compared with the BR-SVM, CPLST, BR-Ridge, BR-DR, PLST, and BR-DT models regarding the AP, EM, HS, Macro, and Micro F1 metrics. [ABSTRACT FROM AUTHOR]

The article examines whether expert knowledge improves the estimation results of real estate mass appraisal models. Six econometric models were compared: OLS, mixed, the Bayesian model, the Inequality Restricted Least Squares (IRLS) model, ridge and LASSO regression (with regularization). In three of the models (mixed, Bayesian, and IRLS) prior knowledge was applied. In mixed and Bayesian models priors took the form of intervals for model parameters. In IRLS, restrictions in the form of inequalities were applied. In the empirical example mass appraisal models were applied in the valuation of undeveloped land for residential purposes. Models with prior knowledge turned out to be the best with regard to the consistency of estimates with theory. Also, prediction accuracy was better for models with prior knowledge. In the case of low quality data expert knowledge might significantly improve estimation results of real estate mass appraisal econometric models. [ABSTRACT FROM AUTHOR]

Instead of picking up a single ridge parameter in ridge regression, this paper considers a frequentist model averaging approach to appropriately combine the set of ridge estimators with different ridge parameters, when the response is randomly right censored. Within this context, we propose a weighted least squares ridge estimation for unknown regression parameter. A new Mallows-type weight choice criterion is then developed to allocate model weights, where the unknown distribution function of the censoring random variable is replaced by the Kaplan–Meier estimator and the covariance matrix of random errors is substituted by its averaging estimator. Under some mild conditions, we show that when the fitting model is misspecified, the resulting model averaging estimator achieves optimality in terms of minimizing the loss function. Whereas, when the fitting model is correctly specified, the model averaging estimator of the regression parameter is root-n consistent. Additionally, for the weight vector which is obtained by minimizing the new criterion, we establish its rate of convergence to the infeasible optimal weight vector. Simulation results show that our method is better than some existing methods. A real dataset is analyzed for illustration as well. [ABSTRACT FROM AUTHOR]

We employ an extended Stochastic Impacts by Regression on Population, Affluence and Technology (STIRPAT) model combined with the environmental Kuznets curve and machine learning algorithms, including ridge and lasso regression, to investigate the impact of institutions on carbon emissions in a sample of 22 European Union countries over 2002 to 2020. Splitting the sample into two: those with weak and strong institutions, we find that the results differ between the two groups. Our results suggest that changes in institutional quality have a limited impact on carbon emissions. Government effectiveness leads to an increase in emissions in the European Union countries with stronger institutions, whereas voice and accountability lead to a fall in emissions. In the group with weaker institutions, political stability and the control of corruption reduce carbon emissions. Our findings indicate that variables such as population density, urbanization and energy consumption are more important determinants of carbon emissions in the European Union compared to institutional governance. The results suggest the need for coordinated and consistent policies that are aligned with climate targets for the European Union as a whole. [ABSTRACT FROM AUTHOR]

This work is concerned with the properties of the ridge regression where the number of predictors p is proportional to the sample size n. Asymptotic properties of the means square error (MSE) of the estimated mean vector using ridge regression is investigated when the design matrix X may be non-random or random. Approximate asymptotic expression of the MSE is derived under fairly general conditions on the decay rate of the eigenvalues of X T X when the design matrix is nonrandom. The value of the optimal MSE provides conditions under which the ridge regression is a suitable method for estimating the mean vector. In the random design case, similar results are obtained when the eigenvalues of E [ X T X ] satisfy a similar decay condition as in the non-random case. [ABSTRACT FROM AUTHOR]

Vegetation is an important link between land, atmosphere, and water, making its changes of great significance. However, existing research has predominantly focused on long-term vegetation changes, neglecting the intra-annual variations of vegetation. Hence, this study is based on the Enhanced Vegetation Index (EVI) data from 2000 to 2022, with a time step of 16 days, to analyze the intra-annual patterns of vegetation changes in China. The average intra-annual EVI values for each municipal-level administrative region were calculated, and the time-series k-means clustering algorithm was employed to divide these regions, exploring the spatial variations in China's intra-annual vegetation changes. Finally, the ridge regression and random forest methods were utilized to assess the drivers of intra-annual vegetation changes. The results showed that: (1) China's vegetation status exhibits a notable intra-annual variation pattern of "high in summer and low in winter," and the changes are more pronounced in the northern regions than in the southern regions; (2) the intra-annual vegetation changes exhibit remarkable regional disparities, and China can be optimally clustered into four distinct clusters, which align well with China's temperature and precipitation zones; and (3) the intra-annual vegetation changes demonstrate significant correlations with meteorological factors such as dew point temperature, precipitation, maximum temperature, and sea-level pressure. In conclusion, our study reveals the characteristics, spatial patterns and driving forces of intra-annual vegetation changes in China, which contribute to explaining ecosystem response mechanisms, providing valuable insights for ecological research and the formulation of ecological conservation and management strategies. [ABSTRACT FROM AUTHOR]

Global warming upsets the environmental balance and leads to more frequent and severe climatic events. These extreme events include floods, droughts, and heatwaves. These widespread extreme events disrupt various sectors of ecosystems directly. However, among all these events, drought is one of the most prolonged climatic events that significantly destroys the ecosystem. Therefore, accurate and efficient assessment of droughts is necessary to mitigate their detrimental impacts. In recent years, several drought indices based on global climate models (GCMs) of Coupled Model Intercomparison Project Phase 6 (CMIP6) have been proposed to quantify and monitor droughts. However, each index has its advantages and limitations. As each index ensembles different models by using different statistical approaches, it is well known that the margin of error is always a part of statistics. Therefore, this study proposed a new drought index to reduce the uncertainty involved in the assessment of droughts. The proposed index named the Ridge Ensemble Standardized Drought Index (RESDI) is based on the innovative ensemble approach termed ridge parameters and distance-based weighting (RDW) scheme. And the development of this RDW scheme is based on two types of methods i.e., ridge regression and divergence-based method. In this research, we ensemble 18 different GCMs of CMIP6 using the RDW scheme. A comparative analysis of the RDW scheme is performed against the simple model average (SMA) and Bayesian model averaging (BMA) schemes at 32 locations on the Tibetan plateau. The comparison revealed that RDW has less mean absolute error (MAE) and root-mean-square error (RMSE). Therefore, the developed RESDI based on RDW is used to project drought properties under three distinct shared socioeconomic pathway (SSP) scenarios: SSP1-2.6, SSP2-4.5, and SSP5-8.5, across seven different time scales (1, 3, 7, 9, 12, 24, and 48). The projected data is then standardized by using the K-components Gaussian mixture model (K-CGMM). In addition, the study employs steady-state probabilities (SSPs) to determine the long-term behavior of drought. The outcome of this research shows that "normal drought (ND)" has the highest probability of occurrence under all scenarios and time scales. [ABSTRACT FROM AUTHOR]

This paper provides evidence on the use of Random Regression Forests (RRF) for optimal lag selection. Using an extended sample of 144 data series, of various data types with different frequencies and sample sizes, we perform optimal lag selection using RRF and compare the results with seven "traditional" information criteria as well as with three other machine learning approaches. We show that the different information criteria produce differing outcomes in terms of optimal lag selection. To quantify performance, we compare the forecast errors on autoregressive models using the optimal lags selected by the criteria and demonstrate that RRF outperforms other approaches. We provide suggestions to researchers as to which approach to use, under different combinations of data type/data frequency and data type/sample size. [ABSTRACT FROM AUTHOR]

The determinants of mortgage default have been an area of rising interest since the 2008 recession. There are two distinguishing features of mortgage default analysis. First, predictor variables are often only recorded at origination. However, many important variables such as credit scores vary over time. Second, there are omitted variables (such as borrower's income and job security). If omitted variables are correlated with included regressors or if only origination values are used in a dynamic model, then biases may be present in econometric models for default risk. Our focus is to develop a ridge regression model to impute the dynamics of time-varying predictors and to capture unobserved borrower heterogeneity. The model is evaluated using cross-validation, and the relevant parameters are tuned to maximize out-of-sample predictive performance. After allowing for imputed dynamics and borrower heterogeneity, we find that the loan-to-value ratio becomes a larger signal of default risk and that credit scores as well as full documentation become smaller signals of default risk. These changes primarily are driven by imputing static variables, rather than dynamics, and may pertain to either omitted liquidity factors or strategic factors. [ABSTRACT FROM AUTHOR]

Traditional tracking algorithms based on correlation filtering usually use a filter based on Gaussian-like distribution to highlight the information of the target and weaken the interference of the background on the target. The characteristic of such method is that it highlights the importance of the target center, but it weakens the marginal information of the target. To overcome this problem, this paper proposes a new filter based on power-law probability map and proposes a new tracking algorithm based on the power-law probability map filter and the technology of ridge regression. The filter can screens the overall information of the target effectively, including the marginal information of the target. Firstly, a target block is extracted from the actual image at time t according to the position of the target at time t - 1 , and then a probability map is generated from the target block. Next, the power-law probability map filter is generated by several image processing technologies. On this basis, two parts of HOG features are extracted from the target block, and then these two parts HOG features are filtered and are combined with the ridge regression model to locate and track the target. Finally, a large number of experiments show that our proposed tracking algorithm achieves competitive performance compared with the state-of-the-art tracking algorithms. [ABSTRACT FROM AUTHOR]

This study considers the ℓ 2 -regularized error variance estimator, an effective tool for non-sparse linear models. Specifically, this study investigates previously unexplored theoretical properties of this estimator, particularly its asymptotic bias in high dimensional settings where the number of variables increases with increase in the number of observations. This study proves that the estimator is asymptotically unbiased in large-sample settings ( lim p / n < 1 ), while it is asymptotically biased in general ultra-high dimensional settings ( lim p / n > 1 ). Specifically, the asymptotic bias of the ℓ 2 -regularized estimator in ultra-high dimensional settings is accurately derived where the covariates are independent. [ABSTRACT FROM AUTHOR]

In alignment with China's "dual carbon" goals and its quest to build an ecological civilization, this study scrutinizes the carbon emissions derived from consumer lifestyles, with a particular focus on Beijing, a high-consumption urban metropolis. Utilizing the expanded STIRPAT model and ridge regression, factors such as permanent population, per capita consumption expenditure, energy intensity, energy structure, and consumption structure are examined to evaluate their impact on lifestyle-associated carbon emissions. A scenario analysis is also conducted to project future carbon emissions in Beijing. From 2010 to 2020, there was an overall upward trend in lifestyle-associated carbon emissions, up to a maximum of 87.8260 million tons. Indirect consumption-related carbon emissions, particularly those associated with residential and transportation-related consumption, constituted the primary sources. The most influential factors on carbon emissions were found to be the consumption structure. Notably, adopting a low-carbon consumption mindset and an optimized consumption structure could foster significant carbon reduction. Projections suggest that by 2035, carbon emissions due to residents' consumption could decline by 39.72% under a low-carbon consumption scenario and by 48.74% under a coordinated development scenario. Future efforts should prioritize promoting green, low-carbon living, refining consumption structure and practices, curbing excessive housing consumption, improving energy structure, and raising technological and energy efficiency standards. [ABSTRACT FROM AUTHOR]

The total processing error of CNC machine tools essentially comprises geometric errors and thermal errors. Therefore, reducing the influence of thermal errors is necessary. In this study, 13 temperature sensors were utilized to measure temperature variations of heat sources on a machine. These sensors work in conjunction with a non-contact optical measurement system to measure the positioning offset error of a rotating shaft. This study set α value to be the maximum allowed ridge parameter and named the method Critical–α, allowing an appropriate ridge parameter for use with a ridge regression model to be quickly selected, and integrated into a backward elimination procedure to achieve ridge regression thermal error compensation modeling. The study considered three methods for selecting temperature variable combinations. The first method requires the use of all sensors, the second method selects the combination with the minimum mean-square error, and the third method considers the effect of diminishing returns. The ridge regression method, which considers the diminishing returns effect, is known as the "R–DR model." The R–DR model is applied to the CNC machine used in this study to reduce the maximum peak-to-peak error on the Y-axis from 54.41 to 13.94 µm using only three temperature sensors, and on the Z-axis from 73.59 to 10.12 µm using four temperature sensors. Therefore, the R–DR model has two advantages: high precision (post-compensation peak-to-peak thermal error of less than 14 µm) and fewer temperature sensors, thereby allowing the thermal error compensation modeling method to demonstrate high engineering applicability and accuracy. [ABSTRACT FROM AUTHOR]

Conducting a precise landslide susceptibility assessment at the slope scale is challenging due to complex parameters and limited historical hazards data. This paper proposes a susceptibility assessment model based on the frequency ratio (FR) coupled with multiple regression analysis to solve these problems. Assessment factors were identified through field surveys and remote sensing interpretation. We utilized multiple methods, such as logistic regression (LR), partial least squares regression (PLSR), ridge regression (RR), stepwise regression (SR), and discriminant analysis (DA), and established five coupled models (FR-LR, FR-PLSR, FR-RR, FR-SR, and FR-DA). The reliability of the susceptibility assessment results was systematically verified. The results showed the following: (1) for the distribution of the susceptibility index, FR-PLSR, FR-RR, and FR-DA are close to the standard normal distribution; (2) for the prediction accuracy, the AUC indexes of the five models are relatively large, ranging from 0.86 to 0.87; and (3) for the susceptibility zoning reliability, FR-PLSR and FR-RR showed better performance, with a relatively smaller area proportion and larger susceptibility intensity for high susceptibility regions. Finally, the coupled models FR-PLSR and FR-RR were recommended. The areas of slope units categorized as high-, medium-, low-, and extra low-susceptibility are 16.14 km2, 73.85 km2, 98.96 km2, and 39.49 km2, respectively, accounting for 7.06%, 32.33%, 43.32%, and 17.29% of the study area. This work enriches the theoretical understanding of the susceptibility assessment models with limited historical hazards data and provides an effective coupled model for high-accuracy landslide assessment and prevention at the slope scale. [ABSTRACT FROM AUTHOR]

The government's development of eco-environmental policies can have a scientific foundation thanks to the fine particulate matter (PM2.5) medium- and long-term change forecast. This study develops a STRIPAT-Scenario analysis framework employing panel data from 11 cities in Zhejiang Province between 2006 and 2020 to predict the changing trend of PM2.5 concentrations under five alternative scenarios. The results reveal that: (1) urbanization development (P), economic development (A), technological innovation investment (T) and environmental regulation intensity have a significant inhibitory effect on PM2.5 concentration in Zhejiang Province, while industrial structure, industrial energy consumption and the number of motor vehicles (TR) have a significant increase on PM2.5 concentration. (2) Under any scenario, the PM2.5 concentration of 11 cities in Zhejiang Province can reach the constraint target set in the 14th Five-Year plan. The improvement in urban PM2.5 quality is most obviously impacted by the high-quality development scenario (S4). (3) Toward 2035, PM2.5 concentrations of 11 cities in Zhejiang Province can reach the National Class I level standard in most scenario models, among which Hangzhou, Jiaxing and Shaoxing are under high pressure to reduce emissions and are the key areas for PM2.5 management in Zhejiang Province. However, most cities cannot reach the 10 μg/m3 limit of WHO's AQG2005 version. Finally, this study makes recommendations for reducing PM2.5 in terms of enhancing industrial structure and funding science and technology innovation. [ABSTRACT FROM AUTHOR]

We run a 'horse race' among popular forecasting methods, including machine learning (ML) and deep learning (DL) methods, that are employed to forecast U.S. GDP growth. Given the unstable nature of GDP growth data, we implement a recursive forecasting strategy to calculate the out-of-sample performance metrics of forecasts for multiple subperiods. We use three sets of predictors: a large set of 224 predictors [of U.S. GDP growth] taken from a large quarterly macroeconomic database (namely, FRED-QD), a small set of nine strong predictors selected from the large set, and another small set including these nine strong predictors together with a high-frequency business condition index. We then obtain the following three main findings: (1) when forecasting with a large number of predictors with mixed predictive power, density-based ML methods (such as bagging, boosting, or neural networks) can somewhat outperform sparsity-based methods (such as Lasso) for short-horizon forecast, but it is not easy to distinguish the performance of these two types of methods for long-horizon forecast; (2) density-based ML methods tend to perform better with a large set of predictors than with a small subset of strong predictors, especially when it comes to shorter horizon forecast; and (3) parsimonious models using a strong high-frequency predictor can outperform other sophisticated ML and DL models using a large number of low-frequency predictors at least for long-horizon forecast, highlighting the important role of predictors in economic forecasting. We also find that ensemble ML methods (which are the special cases of density-based ML methods) can outperform popular DL methods. [ABSTRACT FROM AUTHOR]

Ridge regression estimators can be interpreted as a Bayesian posterior mean (or mode) when the regression coefficients follow multivariate normal prior. However, the multivariate normal prior may not give efficient posterior estimates for regression coefficients, especially in the presence of interaction terms. In this paper, the vine copula-based priors are proposed for Bayesian ridge estimators under the Cox proportional hazards model. The semiparametric Cox models are built on the posterior density under two likelihoods: Cox's partial likelihood and the full likelihood under the gamma process prior. The simulations show that the full likelihood is generally more efficient and stable for estimating regression coefficients than the partial likelihood. We also show via simulations and a data example that the Archimedean copula priors (the Clayton and Gumbel copula) are superior to the multivariate normal prior and the Gaussian copula prior. [ABSTRACT FROM AUTHOR]

We propose a fast algorithm for computing the entire ridge regression regularization path in nearly linear time. Our method constructs a basis on which the solution of ridge regression can be computed instantly for any value of the regularization parameter. Consequently, linear models can be tuned via cross-validation or other risk estimation strategies with substantially better efficiency. The algorithm is based on iteratively sketching the Krylov subspace with a binomial decomposition over the regularization path. We provide a convergence analysis with various sketching matrices and show that it improves the state-of-the-art computational complexity. We also provide a technique to adaptively estimate the sketching dimension. This algorithm works for both the over-determined and under-determined problems. We also provide an extension for matrix-valued ridge regression. The numerical results on real medium and large-scale ridge regression tasks illustrate the effectiveness of the proposed method compared to standard baselines which require super-linear computational time. [ABSTRACT FROM AUTHOR]

Rice is one of the most important cereal foods not only for India but also for the world. The production of crop depends upon the favorable climatic conditions. Farmers' access to more accurate data on crop yields in various climate conditions can aid in crucial agronomic and crop selection decisions. Taking this into account, the motive of the present research was to find the best method of predicting rice crop yield in seven important rice producing districts of Uttarakhand, namely Udham Singh Nagar, Nainital, Haridwar, Dehradun, Champawat, Tehri-Garhwal, and Pauri Garhwal. Data on the weather variables for the crop-growing season (27th to 44th SMW) for 19 years was gathered from the respective district and the NASA power website, while rice production data for the research period was gathered from the Directorate of Economics and Statistics, Ministry of Agriculture and Farmers Welfare. Stepwise multiple linear regression (SMLR), least absolute shrinkage and selection operator (LASSO), ridge regression, elastic net (ELNET), and artificial neural network (ANN) were employed for the model's development. The 80% data of the total datasets was utilized to calibrate the models, while the remaining 20% data was allocated for the model validation. On examining these models, LASSO was found to be the finest performing model followed by ELNET, while SMLR was the worst performing model during calibration stage. During validation stage, ANN performed better for Champawat, Dehradun, Haridwar, Pauri Garhwal, and Udham Singh Nagar. The performance of ELENT and LASSO was found to be best for Nainital and Tehri Garhwal, respectively. The performance of ridge regression and SMLR were found to be poor as compared to the other models for the rice yield forecasting. [ABSTRACT FROM AUTHOR]

For solving large-scale ridge regression problem, two adaptive block coordinate descent methods based on the greedy criterion are proposed. The first one called adaptive block Kaczmarz (ABK) method is designed for the underdetermined system while the second one called adaptive block Gauss–Seidel (ABGS) method is designed for the overdetermined system, where both of them do not need to predetermine block pavings or solve any subsystems. The convergence theories are established, and the numerical experiments are conducted. Numerical results illustrate that both ABK and ABGS significantly outperform the state-of-art methods especially in the sense of computing times. [ABSTRACT FROM AUTHOR]

Knowledge of causal influences and the direction of causality that exist between system variables are very important for successful policy design and implementation. This research explores the trivariate causality between economic growth, energy consumption (crude oil), and carbon emissions in India in a multivariate framework with a generalized method of moments (GMM) estimation. The ridge regression estimation is utilized to check the robustness of the GMM model. The GMM results show unidirectional causality from capital stock to economic growth, disposable income to energy consumption, and urbanization to carbon emissions. The study reveals bidirectional causality between economic growth and energy consumption, economic growth and carbon emissions, and energy consumption and carbon emissions. The causality from energy consumption to economic growth and from energy consumption to emissions proves to be a trade-off, as any energy conservation policy toward mitigating emissions can frustrate the economic growth of the country. The feedback causality between economic growth and carbon emissions will result in a negative externality on future investments for having a cleaner environment leading to further slowing down of economic growth. A 1% increase in GDP increases carbon emissions by 1.773%, while a 1% increase in disposable income increases energy consumption by 0.065%. Moreover, a 1% increase in urbanization increases carbon emissions by 0.524%. The counterintuitive feedback causation from carbon emissions (1% increase) to energy consumption (0.950% increase) confirms the existence of a “direct rebound effect.” The study also found empirical evidence for the existence of the Environmental Kuznets curve in the Indian context. Finally, some useful policy implications are discussed. [ABSTRACT FROM AUTHOR]

The agriculture sector is one of the leading emitters of greenhouse gases in Bangladesh, owing to increasing mechanization, changing population patterns and increasing cultivation of irrigation intensive crops like rice. The objective of this research is to analyze how population trends, energy use and land use practices impact the emissions of three greenhouse gases from the agriculture sector in Bangladesh. The gases studied are carbon dioxide, methane and nitrous oxide. The Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) model and ridge regression are used to analyze the drivers of emissions covering the period from 1990 to 2014. Explanatory factors of emissions are the total and rural population, affluence, urbanization, fertilizer intensity and quantity, carbon and energy intensity, irrigation, rice cultivation, cultivated land and crop yield. The findings reveal that the country's total population has a negative effect, and the rural population has a negative, nonlinear impact on the emissions of methane. Affluence affects emissions of all the gases. The energy intensity and carbon intensity of agriculture increase carbon dioxide emissions. The cultivated land area, rice cultivation quantity and crop yield increase methane emissions, while irrigated land area decreases it. Rural population, total population and urbanization have a positive linear effect on carbon dioxide and nitrous oxide emissions. Fertilizer quantity and intensity increase nitrous oxide emissions. The findings imply that increasing agricultural mechanization should be based on clean energy, and land management should be regulated to enable the country to meet its Nationally Determined Contribution (NDC) targets as well as the targets of Sustainable Development Goal (SDG) 7 of increasing the share of clean energy. [ABSTRACT FROM AUTHOR]

Homomorphic encryption (HE) has recently attracted considerable attention as a key solution for privacy-preserving machine learning because HE can apply to various areas that require to delegate outsourcing computations of user's data. Nevertheless, its computational inefficiency still hinders its wider application. In this study, we propose an alternative to bridge the gap between the privacy and efficiency of HE by encrypting only a small amount of private information. We first derive an exact solution to HE-friendly ridge regression with multiple private variables, while linearly reducing the computational complexity of this algorithm over the number of variables. The proposed method has the advantage that it can be implemented using any HE scheme. Moreover, we propose an adversarial perturbation method that can prevent potential attacks on private variables, which have rarely been explored in HE-based machine learning studies. An extensive experiment on real-world benchmarking datasets supports the effectiveness of our method. [ABSTRACT FROM AUTHOR]

The impact of global greenhouse gas emissions is increasingly serious, and the development of green low-carbon circular economy has become an inevitable trend for the development of all countries in the world. To achieve emission peak and carbon neutrality is the primary goal of energy conservation and emission reduction. As the core province in central China, Hubei Province is under prominent pressure of carbon emission reduction. In this paper, the future development trend of carbon emissions is analyzed, and the emission peak value and carbon peak time in Hubei Province is predicted. Firstly, the generalized Divisia index method (GDIM) model is proposed to determine the main influencing factors of carbon emissions in Hubei Province. Secondly, based on the main influencing factors identified, a novel STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) extended model with ridge regression is established to predict carbon emissions. Thirdly, the scenario analysis method is used to set the variables of the STIRPAT extended model and to predict the emission peak value and carbon peak time in Hubei Province. The results show that Hubei Province's carbon emissions peaked first in 2025, with a peak value of 361.81 million tons. Finally, according to the prediction results, the corresponding suggestions on carbon emission reduction are provided in three aspects of industrial structure, energy structure, and urbanization, so as to help government establish a green, low-carbon, and circular development economic system and achieve the industry's cleaner production and sustainable development of society. [ABSTRACT FROM AUTHOR]

In order to cope with global warming, China has put forward the "30 · 60" plan. We take Henan Province as an example to explore the accessibility of the plan. Tapio decoupling model is used to discuss the relationship between carbon emissions and economy in Henan Province. The influence factors of carbon emissions in Henan Province were studied by using STIRPAT extended model and ridge regression method, and the carbon emission prediction equation was obtained. On this basis, the standard development scenario, low-carbon development scenario, and high-speed development scenario are set according to the economic development model to analyze and predict the carbon emissions of Henan Province from 2020 to 2040. The results show that energy intensity effect and energy structure effect can promote the optimization of the relationship between economy and carbon emissions in Henan Province. Energy structure and carbon emission intensity have a significant negative impact on carbon emissions, while industrial structure has a significant positive impact on carbon emissions. Henan Province can achieve the "carbon peak" goal by 2030 years under the standard and low-carbon development scenario, but it cannot achieve this goal under the high-speed development scenario. Therefore, in order to achieve the goals of "carbon peaking" and "carbon neutralization" as scheduled, Henan Province must adjust its industrial structure, optimize its energy consumption structure, improve energy efficiency, and reduce energy intensity. [ABSTRACT FROM AUTHOR]

Forensic diagnosis of sudden cardiac death (SCD) is an extremely important part of routine forensic practice. The present study aimed to develop and validate nomograms for predicting the probability of SCD with special regards to ischemic heart disease-induced SCD (IHD-induced SCD) based on multiple autopsy variables. A total of 3322 cases, were enrolled and randomly assigned into a training cohort (n = 2325) and a validation cohort (n = 997), respectively. Prediction models of SCD and IHD-induced SCD were developed through multivariable logistic regression based on variables selected by LASSO regression or ridge regression, and prediction model with higher area under the curve (AUC) of the receiver operating characteristic (ROC) curve in the validation cohort was used to establish nomograms. For SCD prediction, discrimination of the nomogram was determined based on the ROC with AUC of 0.751 (95% CI, 0.726–0.775) and 0.735 (95% CI, 0.696–0.774) in the training cohort and validation cohort respectively. The AUC of IHD-induced SCD prediction nomogram in the training cohort and validation cohort were 0.742 (95% CI, 0.716–0.768) and 0.738 (95% CI, 0.698–0.777). To facilitate the use of nomograms in routine casework in forensic practice, web calculators (https://forensic.shinyapps.io/Forensic_SCD/, https://forensic.shinyapps.io/Forensic_IHDinducedSCD/) were constructed. In conclusion, the present study developed and validated simple and practical nomograms for predicting the probability of SCD and IHD-induced SCD based on multiple autopsy variables. The nomograms have certain efficiency for discrimination and calibration to provide a novel approach to diagnose cause of death, and may become a valuable tool in forensic practice.  [ABSTRACT FROM AUTHOR]

Nowadays, smart health technologies are used in different life and environmental areas, such as smart life, healthcare, cognitive smart cities, and social systems. Intelligent, reliable, and ubiquitous healthcare systems are a part of the modern developing technology that should be more seriously considered. Data collection through different ways, such as the Internet of things (IoT)-assisted sensors, enables physicians to predict, prevent and treat diseases. Machine Learning (ML) algorithms may lead to higher accuracy in medical diagnosis/prognosis based on health data provided by the sensors to help physicians in tracking symptom significance and treatment steps. In this study, we applied four ML methods to the data on Parkinson's disease to assess the methods' performance and identify the essential features that may be used to predict the total Unified Parkinson's disease Rating Scale (UPDRS). Since accessibility and high-performance decision-making are so vital for updating physicians and supporting IoT nodes (e.g., wearable sensors), all the data is stored, updated as rule-based, and protected in the cloud. Moreover, by assigning more computational equipment and memory in use, cloud computing makes it possible to reduce the time complexity of the training phase of ML algorithms in the cases we want to create a complete structure of cloud/edge architecture. In this situation, it is possible to investigate the approaches with varying iterations without concern for system configuration, temporal complexity, and real-time performance. Analyzing the coefficient of determination and Mean Square Error (MSE) reveals that the outcomes of the applied methods are mostly at an acceptable performance level. Moreover, the algorithm's estimated weight indicates that Motor UPDRS is the most significant predictor of Total UPDRS. [ABSTRACT FROM AUTHOR]

The electricity sector is the biggest emitter of greenhouse gases in Bangladesh, and the country has the target to reduce these emissions and follow an environmentally sustainable development strategy. In this study, we explore how population growth, affluence and energy technology factors affect the emissions of carbon dioxide, methane and nitrous oxide emissions from the electricity sector in Bangladesh. This study employs an extended stochastic impacts by regression on population, affluence and technology model, using ridge regression for the analysis, covering a period from 1995 to 2014. The study also examines the presence of the environmental Kuznets curve (EKC) with respect to greenhouse gas emissions. We find that the chief drivers of carbon dioxide, methane and nitrous oxide emissions are population, affluence, urbanization and the carbon intensity of fuels. Energy intensity affects carbon dioxide and methane emissions. Transmission losses do not have a significant positive influence on emissions of any of the gases except carbon dioxide. Our results also show that there is no evidence of the EKC with respect to greenhouse gases in the electricity sector. Our study suggests that Bangladesh should have policies promoting a cleaner energy mix, reduce the use of fossil fuels, and increase energy efficiency technologies to counter the polluting effects of a growing power sector. [ABSTRACT FROM AUTHOR]

Spectral clustering is a recent proposed popular method for unsupervised image clustering problem, whose performance relies on a well-designed affinity matrix that measures the similarity among data. Nevertheless, on one hand, most of the existed affinity learning methods have been proposed focus on single-feature data, which would lose some cues from their other useful heterogeneous features. On the other hand, the specific two dimensional structures of visual data are not well preserved when affinity is learned. To address these two problems, in this paper, we propose a novel heterogeneous feature based strategy for image clustering by combining projection learning and affinity learning into a joint learning framework. Inspired by multimodal information integration, heterogeneous modal-based projection learning scheme, that maps each type of heterogeneous feature into the latent subspace in a two-dimensional way, is built with original image and their multi-modal heterogeneous features respectively to preserve more useful structural information. Then, the ensembling projected data of each modal are jointly used to establish an affinity learning model based on the shared ridge regression that considers cues from all modals simultaneously, which can greatly improve the similarity metric performance. Moreover, the graph-based constraint on affinity matrix is also adopt to capture the extra nonlinear structures of the data, which provides help for keeping their neighbor relationships. An iterative numerical scheme is designed to solve the objective function of our proposed framework and ensure convergence. Extensive experimental results on several public datasets verify the advancement and effectiveness of our proposed method among comparison methods. [ABSTRACT FROM AUTHOR]

Ridge regression is a widely used method to mitigate the multicollinearly problem often arising in multiple linear regression. It is well known that the ridge regression estimator can be derived from the Bayesian framework by the posterior mode under a multivariate normal prior. However, the ridge regression model with a copula-based multivariate prior model has not been employed in the Bayesian framework. Motivated by the multicollinearly problem due to an interaction term, we adopt a vine copula to construct the copula-based joint prior distribution. For selected copulas and hyperparameters, we propose Bayesian ridge estimators and credible intervals for regression coefficients. A simulation study is carried out to compare the performance of four different priors (the Clayton, Gumbel, and Gaussian copula priors, and the tri-variate normal prior) on the regression coefficients. Our simulation studies demonstrate that the Archimedean (Clayton and Gumbel) copula priors give more accurate estimates in the presence of multicollinearity compared with the other priors. Finally, a real dataset is analyzed, where the Bayesian ridge estimators and some frequentist estimators are compared. [ABSTRACT FROM AUTHOR]

In this paper, we present a general framework for estimating regression models subject to a user-defined level of fairness. We enforce fairness as a model selection step in which we choose the value of a ridge penalty to control the effect of sensitive attributes. We then estimate the parameters of the model conditional on the chosen penalty value. Our proposal is mathematically simple, with a solution that is partly in closed form and produces estimates of the regression coefficients that are intuitive to interpret as a function of the level of fairness. Furthermore, it is easily extended to generalised linear models, kernelised regression models and other penalties, and it can accommodate multiple definitions of fairness. We compare our approach with the regression model from Komiyama et al. (in: Proceedings of machine learning research. 35th international conference on machine learning (ICML), vol 80, pp 2737–2746, 2018), which implements a provably optimal linear regression model and with the fair models from Zafar et al. (J Mach Learn Res 20:1–42, 2019). We evaluate these approaches empirically on six different data sets, and we find that our proposal provides better goodness of fit and better predictive accuracy for the same level of fairness. In addition, we highlight a source of bias in the original experimental evaluation in Komiyama et al. (in: Proceedings of machine learning research. 35th international conference on machine learning (ICML), vol 80, pp 2737–2746, 2018). [ABSTRACT FROM AUTHOR]

Tunisia is among the developing countries that have taken initiatives to develop renewable energy and strengthen energy efficiency. Moreover, it has considerable potential, especially in the field of wind and solar energy. However, the country is still dependent on fossil fuel energy. In this context, the transition to renewable energy is considered one of the possible solutions to reduce energy dependence and strengthen the economy in general. Therefore, the aim of this research study is to evaluate the role of renewable energy in shaping the energy transition in Tunisia in order to qualify the possibilities of energy transition. Accordingly, we investigate the potential for substitution between the following factors and fuels: Capital, labor, renewable, and non-renewable energy in Tunisia using a translog production function approach. Due to the multicollinearity of the model, the ridge regression method is used to estimate the parameters of the function. The obtained results showed that the possibility of substitution between inputs, especially between renewable and non-renewable energy, can replace fossil fuels with clean energy consumption. Moreover, to maximize the potential of renewable energy in Tunisia, this study recommends that policy makers should take more reliable measures to reflect the exact price of energy through price regulation measures, encourage investment in research and development, and introduce carbon taxes that could accelerate this transition. [ABSTRACT FROM AUTHOR]

Recent work has shown that despite their simplicity, item-based models optimised through ridge regression can attain highly competitive results on collaborative filtering tasks. As these models are analytically computable and thus forgo the need for often expensive iterative optimisation procedures, they have become an attractive choice for practitioners. Computing the closed-form ridge regression solution consists of inverting the Gramian item-item matrix, which is known to be a costly operation that scales poorly with the size of the item catalogue. Because of this bottleneck, the adoption of these methods is restricted to a specific set of problems where the number of items is modest. This can become especially problematic in real-world dynamical environments, where the model needs to keep up with incoming data to combat issues of cold start and concept drift. In this work, we propose Dynamic E A S E R : an algorithm based on the Woodbury matrix identity that incrementally updates an existing regression model when new data arrives, either approximately or exact. By exploiting a widely accepted low-rank assumption for the user-item interaction data, this allows us to target those parts of the resulting model that need updating, and avoid a costly inversion of the entire item-item matrix with every update. We theoretically and empirically show that our newly proposed methods can entail significant efficiency gains in the right settings, broadening the scope of problems for which closed-form models are an appropriate choice. [ABSTRACT FROM AUTHOR]

The perspectives of geography and medicine are integrated to analyze the influence of geographic factors on the reference value of tumor marker CYFRA21-1, and explore the spatial distribution of tumor marker CYFRA21-1 for healthy adults in China; CYFRA21-1 is a tumor marker for adults in different regions of China. The value of medical diagnosis provides a theoretical basis. Based on the measured values of the tumor marker CYFRA21-1 in 24,634 healthy adults in 314 cities, combined with 15 geographical factors, predictive modeling was carried out through spatial autocorrelation, correlation analysis, ridge regression analysis, and support vector machine, and the health of 2322 cities in China was predicted. The reference value of the tumor marker CYFRA21-1 for healthy adults in China was used to generate a geographical distribution map. The results showed that the distribution of reference values of the tumor marker CYFRA21-1 for healthy adults in China showed a higher trend in northern and northwestern areas, and a lower trend in southern areas. The reference value of tumor marker CYFRA21-1 has a certain concentration in the geographical distribution of China, and temperature, precipitation, wind speed, and soil factors will affect the reference value of CYFRA21-1 of healthy adults in China. [ABSTRACT FROM AUTHOR]

China is facing increasing pressure to reduce CO2 emissions from energy consumption. Given this issue, understanding the characteristics, influencing factors, and trends can provide adequate information for decision-makers to solve the CO2 emission problem. This study analyzes the characteristics of CO2 emissions from energy consumption in 30 regions of China from 2005 to 2018 and applies the STIRPAT model to identify the impact of the influencing factors. Combined with the CO2 emission trend in 2030 as predicted by the ARIMA model, the key mitigation regions and strategies reduction have been determined. Results indicate that CO2 emissions have been increasing from 2005 to 2018 in China, thus showing the characteristic of the east being larger than the west spatially. Under the baseline scenario, these emissions will continue to rise in 2030. Carbon emissions intensity is declining, and the gap between provinces with the highest and lowest per capita CO2 emissions is widening. Although per capita GDP is significantly positively correlated with provinces, population is the key factor influencing more provinces, followed by the proportion of the secondary industry and urbanization rate. To achieve low-carbon sustainable development, Shandong, Shanxi, Inner Mongolia, Guangdong, Shaanxi, Xinjiang, and Ningxia are considered the key regions of concern for emission reduction. The heterogeneity of CO2 emission characteristics and influencing factors among regions provides a direction for the development of targeted and differentiated regional emission reduction strategies. [ABSTRACT FROM AUTHOR]

Laser-based directed energy deposition (L-DED) is a rising field in the arena of metal additive manufacturing and has extensive applications in aerospace, medical, and rapid prototyping. The process parameters, such as laser power, scanning speed, and layer thickness, play an important role in controlling and affecting the properties of DED fabricated parts. Nevertheless, both experimental and simulation methods have shown constraints and limited ability to generate accurate and efficient computational predictions on the correlations between the process parameters and the final part quality. In this paper, two data-driven machine learning algorithms, Extreme Gradient Boosting (XGBoost) and Random Forest (RF), were applied to predict the tensile behaviors including yield strength, ultimate tensile strength, and elongation (%) of the stainless steel 316L parts by DED. The results suggest that both models successfully predicted the tensile properties of the fabricated parts. The performance of the proposed methods was evaluated and compared with the Ridge Regression by the root mean squared error (RMSE), relative error (RE), and coefficient of determination (R2). XGBoost outperformed both Ridge Regression and Random Forest in terms of prediction accuracy. [ABSTRACT FROM AUTHOR]

Wind energy is one of the renewable energy sources that has been touted to address the challenges of energy security and environmental degradation. This is only attainable if countries with substantial wind energy potential use it in significant proportion to satisfy their energy needs. One promising sector where wind energy can be employed to actualize this potential is the electricity sector. However, the current reality is that fossil fuels still dominate the energy profiles of most economies of the world, including the advanced economies, with wind renewable energy source accounting for a very small proportion of the energy mix. Germany is one of the few countries that offers promising opportunities in deploying wind energy to its full potentials. This study therefore explores the feasibility of substituting wind energy for nuclear energy and other fossil fuels using Germany as a country of focus. We use the ridge regression procedure to analyse yearly time series data for the German power sector that spans the period 1986 to 2018. With respect to output elasticities of the energy inputs, the results reveal that wind and natural gas have positive output elasticity estimates while the estimates for nuclear and coal are negative. We also found that all the inputs pairs have positive substitution elasticity estimates between them. With respect to wind energy, the highest substitutability estimate occurred with nuclear power which is followed by natural gas and then coal. The study recommended that policies such as granting of tax credit for wind energy technology, reduction in property taxes for wind power facilities, and allocation of fund for research and development (R&D) in wind energy technology are recommended to promote the use of wind energy in the economy. [ABSTRACT FROM AUTHOR]

This paper presents a formalization of steps for identification of echo-state network (ESN)-based models for multiple-input and multiple-output (MIMO) nonlinear dynamic systems models, employing the best practices found in the literature. The ESN is a recurrent neural network capable to model nonlinear dynamics and it can be trained using computationally efficient algorithms. A simplified ESN architecture from the literature for process identification tasks is used and a procedure to create the model and tune the main hyperparameters is proposed. The proposed method is evaluated using both a simulated pH neutralization process and a real refrigeration compressor test rig. For each case study, the procedure to obtain the data series, the model creation, and the influence of each hyperparameter in the model performance are discussed. In both cases, the proposed model architecture presented better results than the linear model and the two nonlinear models used as baseline, namely an extreme learning machine-based Hammerstein model and a long short-term memory network model. The results show that the proposed ESN-based system identification method can be used to obtain MIMO models for nonlinear processes using a computationally efficient training algorithm. For the ESN architecture considered in the case studies, the training time is about 80 s, while a prediction can be obtained in about 15 ms. [ABSTRACT FROM AUTHOR]

Elastic moduli such as Young's modulus (E), Poisson's ratio (v), and bulk modulus (K) are vital to creating geomechanical models for wellbore stability, hydraulic fracturing, sand production, etc. Due to the difficulty of obtaining core samples and performing rock testing, alternatively, wireline measurements can be used to estimate dynamic moduli. However, dynamic moduli are significantly different from elastic moduli due to many factors. In this paper, correlations for three zones (Nahr Umr shale, Zubair shale, and Zubair sandstone) located in southern Iraq were created to estimate static E, K, and ν from dynamic data. Core plugs from the aforementioned three zones alongside wireline measurements for the same sections were acquired. Single-stage triaxial (SST) tests with CT scans were executed for the core plugs. The data were separated into two parts; training (70%), and testing (30%) to ensure the models can be generalized to new data. Regularized ridge regression models were created to estimate static E, K, and ν from dynamic data (wireline measurements). The shrinkage parameter (α) was selected for each model based on an iterative process, where the goal is to ensure having the smallest error. The results showed that all models had testing R2 ranging between 0.92 and 0.997 and consistent with the training results. All models of E, K, and ν were linear besides ν for the Zubair sandstone and shale which were second-degree polynomial. Furthermore, root means squared error (RMSE) and mean absolute error (MAE) were utilized to assess the error of the models. Both RMSE and MAE were consistently low in training and testing without a large discrepancy. Thus, with the regularization of ridge regression and consistent low error during the training and testing, it can be concluded that the proposed models can be generalized to new data and no overfitting can be observed. The proposed models for Nahr Umr shale, Zubair shale, and Zubair sandstone can be utilized to estimate E, K, and ν based on readily available dynamic data which can contribute to creating robust geomechanical models for hydraulic fracturing, sand production, wellbore stability, etc. [ABSTRACT FROM AUTHOR]

There are a variety of settings where vague prior information may be available on the importance of predictors in high-dimensional regression settings. Examples include the ordering on the variables offered by their empirical variances (which is typically discarded through standardisation), the lag of predictors when fitting autoregressive models in time series settings, or the level of missingness of the variables. Whilst such orderings may not match the true importance of variables, we argue that there is little to be lost, and potentially much to be gained, by using them. We propose a simple scheme involving fitting a sequence of models indicated by the ordering. We show that the computational cost for fitting all models when ridge regression is used is no more than for a single fit of ridge regression, and describe a strategy for Lasso regression that makes use of previous fits to greatly speed up fitting the entire sequence of models. We propose to select a final estimator by cross-validation and provide a general result on the quality of the best performing estimator on a test set selected from among a number M of competing estimators in a high-dimensional linear regression setting. Our result requires no sparsity assumptions and shows that only a log M price is incurred compared to the unknown best estimator. We demonstrate the effectiveness of our approach when applied to missing or corrupted data, and in time series settings. An R package is available on github. [ABSTRACT FROM AUTHOR]

Measurement error models are studied extensively in the literature. In these models, when the measurement error variance Σ δ δ is known, the estimating techniques require positive definiteness of the matrix S xx - Σ δ δ , even when this is positive definite, it might be near singular if the number of observations is small. There are alternative estimators discussed in literature when this matrix is not positive definite. In this paper, estimators when the matrix S xx - Σ δ δ is near singular are proposed and it is shown that these estimators are consistent and have the same asymptotic properties as the earlier ones. In addition, we show that our estimators work far better than the earlier estimators in case of small samples and equally good for large samples. [ABSTRACT FROM AUTHOR]

In this paper, we propose a numerical boundary treatment for simulating shock propagation in the fractional KdV-Burgers equation, based on a machine learning strategy. Numerical boundary treatment is particularly challenging due to the nonlinear and global interaction among the fractional diffusion, convection and dispersion. We select a suitable number of boundary points and interior points, and correct the numerical value at the boundary ones by linear combination of those at the interior ones. The large set of parameters are trained from a numerical reference shock solution connecting the end states 1 and 0, by ridge regression. Numerical tests demonstrate that the boundary treatment is effective in reflection suppression, and preserves the correct shock speed even under perturbed initial profile. Furthermore, using the above parameters, we construct a modified boundary treatment to simulate a range of different end states, with effectiveness checked numerically. [ABSTRACT FROM AUTHOR]

Due to the recent worldwide outbreak of COVID-19, there has been an enormous change in our lifestyle and it has a severe impact in different fields like finance, education, business, travel, tourism, economy, etc., in all the affected countries. In this scenario, people must be careful and cautious about the symptoms and should act accordingly. Accurate predictions of different factors, like the end date of the pandemic, duration of lockdown and spreading trend can guide us through the pandemic and precautions can be taken accordingly. Multiple attempts have been made to model the virus transmission, but none of them has investigated it at a global level. The novelty of the proposed work lies here. In this paper, first, authors have analysed spreading of the said disease using data collected from various platforms and then, have presented a predictive mathematical model for fifteen countries from first, second and third world for probable future projections of this pandemic. The prediction can be used by planning commission, healthcare organizations and the government agencies as well for creating suitable arrangements against this pandemic. [ABSTRACT FROM AUTHOR]

Consider the multiple linear regression model Y = β 1 + β 2 x 2 + ⋯ + β p x p + e = x T β + e with sample size n. This paper compares the six shrinkage estimators: forward selection, lasso, partial least squares, principal components regression, lasso variable selection, and ridge regression, with large sample theory and two new prediction intervals that are asymptotically optimal if the estimator β ^ is a consistent estimator of β . Few prediction intervals have been developed for p > n , and they are not asymptotically optimal. For p fixed, the large sample theory for variable selection estimators like forward selection is new, and the theory shows that lasso variable selection is n consistent under much milder conditions than lasso. This paper also simplifies the proofs of the large sample theory for lasso, ridge regression, and elastic net. [ABSTRACT FROM AUTHOR]

In this paper, we proposed a novel regularization and variable selection algorithm called Liu–Lasso extreme learning machine (LL-ELM) in order to deal with the ELM's drawbacks like instability, poor generalizability and underfitting or overfitting due to the selection of inappropriate hidden layer nodes. Liu estimator, which is a statistically biased estimator, is considered in the learning phase of the proposed algorithm with Lasso regression approach. The proposed algorithm is compared with the conventional ELM and its variants including ELM forms based on Liu estimator (Liu-ELM), L 1 -norm (Lasso-ELM), L 2 -norm (Ridge-ELM) and elastic net (Enet-ELM). Convenient selection methods for the determination of tuning parameters for each algorithm have been used in comparisons. The results show that there always exists a d value such that LL-ELM overperforms either Lasso-ELM or Enet-ELM in terms of learning and generalization performance. This improvement in LL-Lasso over Lasso-ELM and Enet-ELM in the sense of testing root mean square error varies up to 27 % depending on the proposed d selection methods. Consequently, LL-ELM can be considered as a competitive algorithm for both regressional and classification tasks because of easy integrability property. [ABSTRACT FROM AUTHOR]