Search

Since the inception of operations research, linear programming has received the attention of researchers in this field due to the many areas of its use. The focus was on the methods used to find the optimal solution for linear models. The direct simplex method, with its three basic stages, begins by writing the linear model in standard form and then finding a basic solution that is improved according to the simplex steps until We get the optimal solution, but we encounter many linear models that do not give us a basic solution after we put it in a standard form, and here we need to solve a rule through which we reach the optimal solution. For these models, researchers and scholars in the field of operations research introduced the simplex method with an artificial basis, which helped to Find the optimal solution for linear models, given the importance of this method and as a complement to the previous research we presented using the concepts of neutrosophic science. In this research, we will reformulate the simplex algorithm with an artificial basis using concepts of neutrosophic science. [ABSTRACT FROM AUTHOR]

Background: Faba bean (Vicia faba L.) is an important crop with high protein content. Tens of thousands of faba bean accessions are available in germplasm collections throughout the world. Morphological characterization of these materials can enrich these collections and aid in the selection of genotypes for use in breeding programs. Results: In this study, 26 morphological characters were analyzed for 61 faba bean landraces and 53 cultivars over two seasons in Izmir, Turkey. The genotypes had high diversity for several yield traits including number of pods per plant, dry seed yield, and 100‐seed weight. Association mapping was conducted for the morphological characters using 651 alleles from 100 simple sequence repeat (SSR) markers and a general linear model based on the Q matrix. A false discovery rate of 0.20 was used to test the significance of marker–trait associations resulting in 75 loci detected for 20 of the morphological characters (p ≤ 0.001). Conclusion: Overall, 44% of the quantitative trait loci (QTLs) were for seed traits, with 24%, 15%, and 17% of QTL identified for vegetative, inflorescence, and pod traits, respectively. The phenotypic data and marker–trait associations generated by this work can help breeding programs in the selection and improvement of faba bean. [ABSTRACT FROM AUTHOR]

The study of time and the response of individuals to contextual changes are crucial aspects in the social sciences, especially when analyzing educational inequalities and parental involvement during early stages of schooling. This work contrasts certain conclusions made by Alonso-Carmona and Martín-Criado (2022) in their critical examination of the general linear models applied to survey data, especially regarding the impact of parental practices and expectations on children's education. In response to the critique made by Andrew Abbott (2001), a detailed examination is offered of the critical postulates related to generalized linear models, while also examining the alternative model proposed by Alonso-Carmona and Martín-Criado: the processual and relational approach, based on the conclusions reached. [ABSTRACT FROM AUTHOR]

Emerging data suggests the COVID-19 crisis exacerbated preexisting, long-documented gender inequities among U.S. faculty in higher education. During the initial Spring 2020 'lockdown' in the U.S., 80 students conveyed their experiences with faculty across 362 courses. We evaluated whether students' reports of faculty supportiveness, accommodations granted, and pandemic-impacted, anticipated grade outcomes differed according to faculty gender via mixed linear models (data on 362 courses were nested within 80 student reporters). Students perceived their women instructors as more supportive, accommodating, and anticipated lesser course grade decreases across the semester than in courses taught by men. Accordingly, we interpret that amidst the 'lockdown' crisis, women faculty earned higher perceived supportiveness and positive student outcomes than their male counterparts. Further, the data likely reflects women faculty's greater conscription into demonstrated care work, despite the coding of such labor as "feminine," thereby rendering such work devalued. To reframe, to the degree that students expect more 'intensive pedagogies,' which invites faculty and administrators to gender disparate demands, such pressures likely translate to 'hidden service' burdens, and correspondingly, less time for career-advancing activities (such as research). Broader implications are discussed, alongside women faculty's documented experiences of acceleration in career and work/family pressures in pandemic-times, which combine to exacerbate long-standing, yet now-amplified penalties, potentially driving a widening, gendered chasm in academic career outcomes. We conclude by offering constructive suggestions to mitigate any discriminatory impacts imposed by students' gendered assessment inputs and expectations. [ABSTRACT FROM AUTHOR]

Although previous research has observed a facilitative influence of the first language (L1) on the acquisition and processing of gender agreement in a second language (L2), particularly in language pairs with similar gender agreement marking, the question of whether knowledge of two languages with gender can confer an additional advantage for L3/Ln (third or additional language) learners has not yet been addressed. The present study aimed to fill this research gap by examining the processing of gender agreement in intermediate and advanced L3/Ln Swedish among two groups of Polish native speakers: 30 L2 English / L3 Swedish learners, and 30 L2 English / L3 German / L4 Swedish learners. Participants were tested by means of a speeded grammaticality judgment task, in which they judged the correctness of indefinite noun phrases that either agreed or did not agree in gender. They also completed an untimed gender assignment task to control for their lexical knowledge of gender. Accuracy and response time data were submitted to Generalized Linear Models. The analysis shows that L4 Swedish learners process noun phrases faster than L3 Swedish learners, but only at the intermediate proficiency level; however, the groups do not differ in their judgment accuracy. This advantage is interpreted in terms of a surface transfer of similar gender agreement marking, which helps the learners automatize gender agreement processes earlier, but does not increase their sensitivity to gender-agreement violations. Moreover, the results accord with previous L2 studies in showing that learners of Swedish as L3/Ln develop sensitivity to ungrammaticality with advancing proficiency and benefit substantially from their gender assignment knowledge in processing gender agreement. Crucially, the present study provides preliminary evidence of a multilingual advantage in processing morphosyntactic features in L3/Ln. [ABSTRACT FROM AUTHOR]

The aim of this study is to prove that the dynamic behavior of tensegrity grids can be controlled. This possibility is very important, especially for deployable structures. The impact the support conditions of the structure on the existence of the immanent characteristics, such as self-stress states and infinitesimal mechanisms, and consequently on the dynamic control, is analyzed. Grids built with the modified Quartex modules are considered. A geometrically non-linear model is used, implemented in an original program written in the Mathematica environment. The results confirm the feasibility of controlling tensegrity structures characterized by the presence of the infinitesimal mechanisms. In the case that the mechanisms do not exist, structures are insensitive to the change of the initial prestress level. The occurrence of mechanisms can be controlled by changing the support conditions of the structure. The obtained results make tensegrity a very promising structural concept, applicable in many areas when conventional solutions are insufficient. [ABSTRACT FROM AUTHOR]

An introduction is presented in which author discusses articles on topics including focuses on Africa which is not only a point of origin to explain diasporic performance traditions, Africa is also a place of return and reinvention disrupting linear models of influence.

Background: Challenges in measuring dynamic scapular orientation limit assessment of scapulothoracic and glenohumeral contributions to shoulder function in children with brachial plexus birth injury (BPBI). Double calibration acromion marker cluster (D-AMC) and linear model approaches have been validated to estimate scapular motion in healthy adults, but neither has been evaluated in BPBI.Research Question: Are the linear model and D-AMC approaches able to accurately estimate scapular orientation in children with BPBI at functional arm postures?Methods: Seventeen children with BPBI positioned their affected limbs in 11 static positions while their segment orientations were measured with motion capture. Linear model and D-AMC estimates of scapular orientation were compared against palpation at six of the static positions with functional relevance to BPBI using a three-way repeat measures ANOVA and a comparison of root mean square errors (RMSE) against literature AMC values for healthy adults.Results: The D-AMC was similar to palpation across all positions and scapular axes while the linear model differed from palpation in a few instances. RMSEs of the D-AMC (3.7-14.8°) and particularly the linear model (4.6-24.8°) were generally at or beyond the upper range of past AMC analyses on healthy adults (1.6-14.2°), especially for more complex, multiplanar arm postures. Despite the D-AMC outperforming the linear model, this approach still produced clinically meaningful (>10°) errors for roughly (12.7-22.5%) of subjects.Significance: Current methods for estimating dynamic scapular orientation remain less than ideal for BPBI. Use of the D-AMC may be appropriate to gain broad insights into general dynamic scapulothoracic and glenohumeral function; however, given their potential for producing clinically meaningful errors, the D-AMC and linear model are not recommended for diagnostic purposes or outcomes assessment on an individual patient basis unless their patient-specific accuracy has been evaluated and confirmed prior to use. [ABSTRACT FROM AUTHOR]

Since World War II, value liberalization has been a worldwide trend, but more recently, there has been a resurgence of conservativism. Modernization and cultural theories have difficulty explaining the shift to liberal or illiberal values, but the political environment, an underestimated contextual factor, could shed light on the mechanisms driving it. This study used hierarchical linear modeling on all six waves of the World Values Survey data (1981–2014) to demonstrate that political freedom helps to explain the rise and fall of liberal attitudes when controlling for societal affluence, inequality, and cultural backgrounds. It finds that political freedom conditions the effects of education: although education is usually considered a liberalizing force, its effect is much weaker in non-free than free societies. The findings remind us of the importance of a free political environment for a robust democracy and point to the complex nature of educational effects, namely the ability of education to socialize people in liberal or illiberal directions depending on the political context and the regime's agenda. [ABSTRACT FROM AUTHOR]

This paper presents injection‐shift‐factor‐based linear modelling for various types of series flexible ac transmission system (FACTS) devices within the DC power flow framework. The presented models allow FACTS devices to be properly integrated in current operation and planning software tools, which is key to harnessing the power flow capabilities provided by FACTS technology. Although recent literature has attempted to develop linear models for FACTS devices, the existing models do not accurately reflect the actual operating range for many FACTS devices. Compared to the existing models, the modelling approach presented here reflects the principle of operation of each type of series FACTS device in adjusting transmission line reactance. Through mathematical derivation, linear constraints for FACTS operation are formulated, which are used to formulate power system operation models. The formulated problems are then analysed through simulation studies on various test systems. The results highlight the significant computational efficiency improvements provided by linear FACTS modelling in DC‐based operation models. [ABSTRACT FROM AUTHOR]

Background: The purpose of this study is to develop and illustrate an optimum variable selection approach using Multiple Linear Regression (MLR) and validation using Multi-Layer Feed Forward Neural Network (MLFF), also known as Multilayer Perceptron Neural Network (MLP) while taking bootstrapping into account. All of the factors specified will be evaluated to determine if they have a meaningful association. Objective:The goal of this study was to create a new method in making predictions for oral health data by combining a few statistical techniques. This would make the model more accurate. Material and Methods:A set of medical data that consists 30 observations was used to develop the methodology. The data descriptions of variables being used in this retrospective research including Creatinine, Fasting Blood Sugar, Haemoglobin A1C, and Urea were evaluated using advanced computational statistical modelling approaches. The medical data are used to test the R syntax that is developed in this study. The statistics for each sample are calculated using a model that incorporates bootstrapping and multiple linear regression techniques. Results:The statistical strategy which combines Bootstrap, MLFF and MLR is better than regular statistical method being used for this type of data. The hybrid model technique's accuracy was increased when the data was separated into training and testing dataset. Four variables are taken into account in this case: fasting blood sugar, creatinine, haemoglobin A1C, and urea. Fasting Blood Sugar (β1: 0.461889; p< 0.05), Creatinine (β3: 0.029761; p< 0.05) and Urea (β2:-0.454766; p< 0.05) are all significant, whereas MLFF and MLRhave the predicted mean square error (PMSE) values of 0.01226 and 0.3531 respectively. Conclusion: The value of PMSE is mainly used to diagnose the performance of MLR and MLFF. The MLFF is used to determine how close predicted values are to real data, while the model that has low PMSE value indicates that the model is accurate. The R syntax for combining Bootstrap, MLR, and MLFF is also included in this research article. In a word, this research establishes the superiority of the hybrid model method. [ABSTRACT FROM AUTHOR]

This study examines the Qur'ān's view towards gender and argues that all three masculine, feminine and egalitarian (gender-inclusive) discourses exist in its text, and that these discourses do not follow a simple and linear model but rather a nonlinear and complex one. It also provides evidence, showing that gender equality in the Qur'ān is achieved in two ways: firstly, through linguistic devices that are devoid of gender distinctions, and secondly, through concurrent use of masculine and feminine gender markers in one context. The masculine discourse is, however, more prominent in the verses of this book and manifests itself to varying degrees in three ways: (1) in some verses, only masculine indicators are used but it can be inferred based on the (Taghlīb) principle that both men and women are the intended addressees; (2) in some verses, rules regarding women are stipulated, whilst men are, in fact, the main addressees and (3) in verses that state rules on reciprocal issues such as divorce and marriage but these rules are addressed to only men. The feminine discourse is also present in the Qur'ān and can be seen in verses that address women. [ABSTRACT FROM AUTHOR]

This paper is concerned with a non‐fragile H∞ state feedback control issue for linear parameter‐varying collaborative adaptive cruise control systems subject to denial‐of‐service attacks. The dynamics of the collaborative adaptive cruise control system is described by a linear model where the deviation of the position and the velocity are selected as the state variables. The attack model is utilized, thereby better reflecting the randomly occurring phenomenon of the denial‐of‐service attacks based on a sequence of binary random variables. The main objective of this note is to develop a non‐fragile state feedback control scheme such that, for denial‐of‐service attacks and possible parameter variations in controller gains, the exponential mean‐square stability and the predefined performance index for the system states are guaranteed simultaneously. By using the matrix analysis techniques and Lyapunov stability theory, sufficient conditions for the desired controller are established and solved based on the solutions to the linear matrix inequality conditions. Finally, a three‐car model is provided to check the feasibility of the designed control scheme. [ABSTRACT FROM AUTHOR]

Although models have been developed for predicting severity of COVID-19 from the medical history of patients, simplified models with good accuracy could be more practical. In this study, we examined utility of simpler models for estimating risk of hospitalization of patients with COVID-19 and mortality of these patients based on demographic characteristics (sex, age, race, median household income based on zip code) and smoking status of 12,347 patients who tested positive at Mass General Brigham centers. The corresponding electronic records were queried (02/26–07/14/2020) to construct derivation and validation cohorts. The derivation cohort was used to fit generalized linear models for estimating risk of hospitalization within 30 days of COVID-19 diagnosis and mortality within approximately 3 months for the hospitalized patients. In the validation cohort, the model resulted in c-statistics of 0.77 [95% CI 0.73–0.80] for hospitalization, and 0.84 [95% CI 0.74–0.94] for mortality among hospitalized patients. Higher risk was associated with older age, male sex, Black ethnicity, lower socioeconomic status, and current/past smoking status. The models can be applied to predict the absolute risks of hospitalization and mortality, and could aid in individualizing the decision making when detailed medical history of patients is not readily available. [ABSTRACT FROM AUTHOR]

Trophic interactions of cave arthropods have been understudied. We used molecular methods (NGS) to decipher the food web in the subterranean ecosystem of the Ardovská Cave (Western Carpathians, Slovakia). We collected five arthropod predators of the species Parasitus loricatus (gamasid mites), Eukoenenia spelaea (palpigrades), Quedius mesomelinus (beetles), and Porrhomma profundum and Centromerus cavernarum (both spiders) and prey belonging to several orders. Various arthropod orders were exploited as prey, and trophic interactions differed among the predators. Linear models were used to compare absolute and relative prey body sizes among the predators. Quedius exploited relatively small prey, while Eukoenenia and Parasitus fed on relatively large prey. Exploitation of eggs or cadavers is discussed. In contrast to previous studies, Eukoenenia was found to be carnivorous. A high proportion of intraguild predation was found in all predators. Intraspecific consumption (most likely cannibalism) was detected only in mites and beetles. Using Pianka's index, the highest trophic niche overlaps were found between Porrhomma and Parasitus and between Centromerus and Eukoenenia, while the lowest niche overlap was found between Parasitus and Quedius. Contrary to what we expected, the high availability of Diptera and Isopoda as a potential prey in the studied system was not corroborated. Our work demonstrates that intraguild diet plays an important role in predators occupying subterranean ecosystems. [ABSTRACT FROM AUTHOR]

Background: During training, service members routinely walk with heavy body borne loads for long periods of time. These loads alter knee biomechanics and may produce jerky knee motions that reportedly increase joint loading and risk of musculoskeletal injury. Yet, it is unknown if service members use jerky knee motions during prolong walking with body borne load.Research Question: To quantify the effects of body borne load and duration of walking on the jerkiness of sagittal and frontal plane knee motion.Methods: Eighteen participants had angular jerk of knee motion quantified while they walked (1.3 m/s) for 60-min with three body borne loads (0, 15, and 30 kg). Peak and cost of angular jerk for sagittal and frontal plane knee motion was quantified and submitted to a repeated measures linear model to test the main effects and interaction of load (0, 15, and 30 kg) and time (0, 15, 30, 45, and 60 min).Results: Body borne load increased peak and cost of angular jerk for sagittal plane knee motion up to 35 % and 110 %, respectively, and frontal plane knee motion up to 20 % and 51 %, respectively (all p<0.001), while jerk cost of frontal plane knee motion (p=0.001) increased 31 % after walking 45 min.Significance: Body borne load produced large (between 20 % and 110 %), incremental increases in angular jerk for both sagittal and frontal plane knee motion; whereas, duration of walking led to a 31 % increase in jerkiness of frontal plane knee motion. Service members who often walking for long periods of time with heavy body borne loads may have greater risk of developing musculoskeletal injury and disease due to large increases in jerky knee motions. [ABSTRACT FROM AUTHOR]

The study of social cognition with functional magnetic resonance imaging (fMRI) affords the use of complex stimulus material. Visual attention to distinct aspects of these stimuli can result in the involvement of remarkably different neural systems. Usually, the influence of gaze on neural signal is either disregarded or dealt with by controlling gaze of participants through instructions or tasks. However, behavioral restrictions like this limit the study's ecological validity. Thus, it would be preferable if participants freely look at the stimuli while their gaze traces are measured. Yet several impediments hamper a combination of fMRI and eye-tracking. In our recent work on neural Theory of Mind processes in alexithymia, we propose a simple way of integrating dwell time on specific stimulus features into general linear models of fMRI data. By parametrically modeling fixations, we were able to distinguish neural processes asssociated with specific stimulus features looked at. Here, I discuss opportunities and obstacles of this approach in more detail. My goal is to motivate a wider use of parametric models — usually implemented in common fMRI software packages — to combine fMRI and eye-tracking data. [ABSTRACT FROM AUTHOR]

The numerical results of mathematical modeling of a two-phase, axisymmetric swirling turbulent flow in the separation zone of a centrifugal separator are presented. Calculations were carried out for various turbulence models: Spalart–Allmaras rotation/curvature correction (SARC) and shear stress transport rotation/curvature correction (SST-RC) linear models, SSG/LRR-RSM-w2012 Reynolds stress nonlinear model, and the new two-fluid model. In the numerical solution the longitudinal-transverse implicit scheme is used where the pressure is coupled with flow velocity fields using the SIMPLEC procedure. The results from comparing numerical calculations against each other and against experimental data are presented. [ABSTRACT FROM AUTHOR]

In the present day, there is an enormous demand in the supply of energy to all users. Since the voltage stability for the distribution system is not steady under any circumstances, therefore, to solve the problem of voltage instability the entire distribution system must simplified properly. Thus, some active support is provided for balancing the demand for electrical energy supply. Hence, the problem of voltage instability has been solved by considering phasor measuring units (PMUs) using antlion optimisation proposed in this study. The cost of PMU has been analysed along with other factors such as observability for effectively placing the PMUs with the identification of the weakest node. The problem is solved by using a non‐linear model and the effectiveness of the proposed approach is tested using IEEE‐15, IEEE‐33, IEEE‐69 Radial Distribution Systems and Croatia distribution grid. The result shows that the suggested algorithm is more effective in getting the optimal PMU placements considering multi‐objective. Hence, the maiden state estimation values for distribution topologies are used. [ABSTRACT FROM AUTHOR]

Introduction: Background: although obesity has been consistently associated with lower 25-OH hydroxyvitamin D-25(OH)D-levels, little is known about the effect of weight change on said 25(OH)D levels. Methods: the present analysis was based on data from the National Health and Nutrition Examination Survey. Percent weight changes were calculated from participant-reported maximum lifetime weight and weight in the past year and 10 years ago. Subsequently, general linear models adjusted for potential confounders were assembled to examine 25(OH)D concentrations across percent weight change categories. Results: a total of 6,237 participants with a mean age of 57.5 (SE, 0.2) years comprised the study sample. After adjustment for potential confounders, subjects who have gained weight ≥ 5 % in the past year and 10 years before had on average 4.5 and 5.1 nmol/L lower 25(OH)D levels than those with a stable weight, respectively. Moreover, this association persisted even among participants with adequate vitamin D intake. Notably, subjects who lost weight > 5 % from their maximum reported weight had significantly higher 25(OH)D levels than those who did not. Conclusion: subjects with a stable weight and those who lost weight ≥ 5% from their maximum reported lifetime weight had significantly higher 25(OH)D concentrations than those who did not. Thus, maintaining a healthy weight over time may be an effective strategy to reach optimal serum 25(OH)D levels. [ABSTRACT FROM AUTHOR]

The husk—the leaf-like outer covering of maize ear—has multiple functions, including protecting the ear from diseases infection and dehydration. In previous studies, we genotyped an association panel of 508 inbred lines genotyped with a total of ~550,000 SNPs (Illumina 50 K SNP Chip and RNA-seq). Genome-Wide Association Studies (GWAS) were conducted on four husk traits: husk length (HL), husk layer number (HN), husk thickness (HT), and husk width (HW). Minimal associations were identified and none of them passed the P-value threshold after a Bonferroni multiple-test correction using a single locus test in framework of mixed linear model. In this study, we doubled the number of SNPs (~1,250,000 in total) by adding GBS and 600 K SNP Chip. GWAS, performed with the recently developed multiple loci model (BLINK), revealed six genetic loci associated with HN and HT above the Bonferroni multiple-test threshold. Five candidate genes were identified based on the linkage disequilibrium with these loci, including GRMZM2G381691 and GRMZM2G012416. These two genes were up-regulation and down-regulation in all husk related tissues, respectively. GRMZM2G381691 associated with HT encoded a CCT domain protein, which expressed higher in tropical than temperate maize. GRMZM2G012416 associated with HN encoded an Armadillo (ARM) repeat protein, which regulated GA signal pathway. These associated SNPs and candidate genes paved a path to understand the genetic architecture of husk in maize. [ABSTRACT FROM AUTHOR]

Human perception involves many cognitive processes, such as memory, attention, and critical thinking. An important cognitive process is memory, which is usually connected with the storing and retrieval of information. Different colors and labeling have diverse physiological effects on humans. Our investigation aimed to determine if a change in color or labeling would have a significant effect on memory span and serial recall. However, our results do not support that coloring and labeling have significant impacts on a subject’s memory. [ABSTRACT FROM AUTHOR]

Computational models are crucial to studying the encoding of individual neurons. Static models are composed of a fixed set of parameters, thus resulting in static encoding properties that do not change under different inputs. Here, we challenge this basic concept which underlies these models. Using generalized linear models, we quantify the encoding and information processing properties of basal ganglia neurons recorded in-vitro. These properties are highly sensitive to the internal state of the neuron due to factors such as dependency on the baseline firing rate. Verification of these experimental results with simulations provides insights into the mechanisms underlying this input-dependent encoding. Thus, static models, which are not context dependent, represent only part of the neuronal encoding capabilities, and are not sufficient to represent the dynamics of a neuron over varying inputs. Input-dependent encoding is crucial for expanding our understanding of neuronal behavior in health and disease and underscores the need for a new generation of dynamic neuronal models. [ABSTRACT FROM AUTHOR]

Gene expression variation can be considered as a phenotype, and it plays an important role in both acclimation and adaption. However, genetic variation of gene expression received much less attention than traditional commercial traits in aquaculture. To estimate the genetic variation and heritability of gene transcription in clam Meretrix petechialis, five Vibrio resistance-related genes were selected for gene expression analysis in the digestive gland, and an animal linear model was used to analyze data from quantitative real-time PCR (qRT-PCR). Among the five genes, BIRC7 showed significant additive genetic variations, the heritability of this gene of 12-month- and 15-month-old clams were 0.84±0.32 and 0.91±0.34, respectively. The heritability of other four genes (Bax, NFIL3, Big-Def, and CTL9) expression were low-to-moderate but not significantly expressed. Additionally, no significant phenotypic and genetic correlations between the BIRC7 transcription trait and body size were detected. This study highlights that certain gene expression variation is heritable and provides a reference for indirect selection of M. petechialis with high Vibrio resistance. [ABSTRACT FROM AUTHOR]

High academic expectation—how far a student expects to get in school—is usually predictive of positive outcomes for a student. Yet less is known about mechanisms behind collective expectation: the proportion of students in a school who expect to pursue further studies. Using urban schools' data from the Education Longitudinal Study 2002, this research examines how collective expectations affect short-term and long-term outcomes, and the predictors of these expectations. Through hierarchical linear models, I find that collective expectations are positively associated with academic outcomes, and that individual expectations remain significant even after controlling for socioeconomic status. I argue that the results illustrate how school structural and economic forces interact with collective expectations in protecting or preventing personal academic attainment. [ABSTRACT FROM AUTHOR]

Objectives: To examine the extent to which adults' and youths' perceptions of collective efficacy align, the shared and unique correlates of adults' and youths' perceptions, and the effects of adults' and youths' perceptions on youths' violence. Method: Descriptive analysis, hierarchical linear modeling, and spatial analysis analyze 1,636 youths from the 2008 Boston Youth Survey and 1,677 adults distributed across 85 neighborhoods from the 2008 Boston Neighborhood Study. Results: Descriptive analysis indicates that Boston adults' and youths' perceptions are largely divergent. Spatial analysis indicates that there is not clustering of either adults' or youths' perceptions across Boston neighborhoods. Multilevel models indicate that adults' and youths' perceptions exhibit divergent etiology: Adults' perceptions align closely with neighborhood collective efficacy and with their own neighborhood perceptions, while youths' perceptions are largely a function of individual differences (race and age) and sociobehavioral factors (social support and educational expectations). Youths' perceptions of collective efficacy, rather than aggregated adults' perceptions of collective efficacy, are inversely associated with youths' violence. Conclusions: Adults' and youths' perceptions of collective efficacy represent distinct constructs. Research should focus on the divergent processes through which adults' and youths' perceptions are generated and the differential effects of adults' and youths' perceptions on youths' behaviors. [ABSTRACT FROM AUTHOR]

This paper develops a methodology for the estimation of the global effect of a treatment from multiple surrogates collected using a longitudinal design. The global effect is modelled as a latent function of time which is allowed to vary for different treatment groups. The latent variable is modelled using low-rank thin plate splines, allowing the shape of the latent profiles to be estimated from data. Identifiability issues are addressed and Bayesian Linear Mixed Model formulation is used which allows for the implementation of WinBUGS. The method is validated through a simulation study and has been implemented on a motivating dataset to measure the effect of exposure to polluted air on lung function. [ABSTRACT FROM AUTHOR]

Purpose: This paper aims to investigate the influence of cognitive style diversity on intra-group relationship conflict and individual-level organizational citizenship behaviors (OCBs). The role of leader-member exchange as a moderating variable is also examined. Design/methodology/approach: The authors used hierarchical linear modeling and hierarchical regression analysis to analyze results from a sample of 344 members from 83 teams nested within 126 departments in six manufacturing organizations in the People's Republic of China. Findings: Results yielded general support for our hypothesized relationships between cognitive style diversity and intra-group relationship conflict. Leader-member exchange was also found to moderate the relationship between these two variables. Contrary to expectations, there were no relationships between these variables and individual-level organizational citizenship behaviors. Originality/value: This research addresses calls from the team diversity and conflict literature to address the understudied area of deep-level cognitive diversity. Second, this study addresses previous calls for more team-level and mixed-level theory and methodology to inform OCB research. Third, this is the first study of group-level cognitive style diversity and the moderating influence of leader-member-exchange and provides valuable insights into ways of mitigating some of the negative effects of cognitive diversity on teams. [ABSTRACT FROM AUTHOR]

Hyperspectral unmixing (HU) plays an important role in the hyperspectral image (HSI) processing and analysis. It aims to decompose mixed pixels into pure spectral signatures and their associated abundances. The autoencoder (AE) framework achieves good performance in HU by automatically learning low-dimensional embedding and reconstructing data. Most of the existing HU methods based on the AE framework can be divided into two categories, one is the pixelto-pixel network focusing on the spectral information of HSI, and the other is the 2DCNN network focusing on the spatial information of HSI. However, HSI contains both spatial and spectral features, it is helpful to add both spectral and spatial information into unmixing methods. Besides, most of the existing unmixing methods are based on linear mixed models (LMM). In LMM, the incident light is assumed to be reflected by each component present in the scene only once prior to collection by the camera sensor, and the observed spectrum is thus a linear combination of the endmembers. However, the real scene is often complex, and due to the existence of spectral variability (SV) and photon nonlinear interaction, the assumption of LMM is often not valid (Zhao et al., 2022). As a result, the LMM hardly makes an accurate unmixing in reality. The excellent potential of neural networks to solve nonlinear problems provides the possibility to establish nonlinear mixed models, which may solve this problem. [ABSTRACT FROM AUTHOR]

It's vital for commercial enterprises to accurately predict demand by utilizing the existing sales data. Such predictive analytics is a crucial part of their decision support systems to increase the profitability of the company. In predictive data analytics, the branch of regression modeling is used to predict a numerical response variable like sale amount. In this category, linear models are simple and easy to interpret yet they permit generalization to very powerful and flexible families of models which are called Generalized linear models (GLM). The generalization potential over simple linear regression can be explained twofold: First, GLM relax the assumption of normally distributed error terms. Moreover, the relationship of the set of predictor variables and the response variable could be represented by a set of link functions rather than the sole choice of the identity function. This work models the sales amount prediction problem through the use of GLM. Unique company sales data are explored and the response variable, sale amount is fitted to the Gamma distribution. Then, inverse link function, which is the canonical one in the case of gamma-distributed response variable is used. The experimental results are compared with the other regression models and the classification algorithms. The model selection is performed via the use of MSE and AIC metrics respectively. The results show that GLM is better than the linear regression. As for the classification algorithms, Random Forest and GLM are the top performers. Moreover, categorization on the predictor variables improves model fitting results significantly. [ABSTRACT FROM AUTHOR]

Abstract The spatial distribution of soil thickness plays a critical role in upland hydrological and biogeochemical processes. Process-based models are thought to have great potentials for predicting soil thickness. However, applications of process-based approaches have been greatly limited by parameter estimation, which typically requires specialized laboratories. In this study, on the basis of field works, an approach for parameter estimation under steady-state assumption is developed using linear and nonlinear soil transport models. Under each model, parameter, namely the ratios between the maximum soil production rate and the diffusion coefficient can be determined. Moreover, the optimal simulation time is derived by a power function of the diffusion coefficient derived under the linear model. The method is then used in a 0.31 ha headwater hillslope in the Hemuqiao catchment, China. According to the suggested method, parameters are determined using soil thickness data collected at 45 different locations on the hillslope, and then soil thickness is predicted. We find the distribution of predicted soil thickness and the root mean square error between the measured and predicted soil thickness for all the soil pits are all the same with the estimated ratios in both models. Furthermore, the derived power function between the optimal simulation time and the diffusion coefficient is verified and the optimal simulation time is determined. Comparisons of simulations and field observations indicate that both models can achieve comparable results, although the procedure developed using the linear model is more suitable for the study area. Furthermore, the linear model is more efficient in parameter estimation, and thus is simpler to employ in practical applications. With our framework, field studies are constrained either on ridges or sideslopes, thus reducing the aimless hunting for sampling sites. This framework provides a useful approach for applying process-based geomorphic models in real catchments. Highlights • Parameter determination limits the application of process-based models. • Theory formulas for parameter estimation under local steady-state is developed. • The approach estimates parameters on the basis of constrained sampling data. • Field works for soil thickness measurements are alleviated guided by the method. [ABSTRACT FROM AUTHOR]

Abstract This paper provides an efficient and rigorous model for 3D seismic analysis of long lined tunnels considering wave passage effects. By taking advantages of the 2.5D model, computation effort is considerably reduced. The water-saturated poroelastic half-space is modelled as two-phase media and soil nonlinearity is considered via the equivalent linear approach. This improved model is realized by a finite element-indirect boundary element (FE-IBE) coupling method. The study reveals that the wave passage effect causes significant longitudinal deformation of the tunnel, and differences in the tunnel seismic responses between the two-phase half-space case and the single-phase half-space case are not negligible. [ABSTRACT FROM AUTHOR]

Abstract This study aimed to examine which relevant variables were associated with adolescents' digital reading achievement using the Programme for International Student Assessment (PISA) 2009 dataset. A three-level hierarchical linear model was employed to identify significant factors predicting digital reading, controlling for print reading ability, and to examine if the extent of the association remained the same across schools and countries. The results confirmed the importance of navigation, metacognitive summary strategies, attitudes toward ICT, and social online reading activities. Female students, on average, underperformed male students when controlling for other variables. However, the associations between these factors and digital reading varied across schools and countries. Navigation appeared universally predictive in digital reading according to the results of the multilevel models. Highlights • Navigation was positively predictive of digital reading ability across schools and countries. • Attitudes toward ICT showed significant predictability in digital reading. • Metacognitive summary strategies still foster digital reading. [ABSTRACT FROM AUTHOR]

Abstract We study generalized partially linear single-index models for longitudinal data in this article. We propose a method to efficiently estimate both the parameters and the nonparametric single-index function in generalized partially linear single-index models when subjects are observed or measured over time. The proposed estimation approach is more flexible and more general in that we can model both categorical response and transformation-necessary response such as heavy-tailed variable with multiple covariates, especially when some covariates are parametrically correlated with the response and the others are nonparametrically correlated with the response. With minimal assumptions, we show that the semiparametric efficiency bound is reached for the parameter estimators. We also show that the asymptotic variance of the single-index function estimator is generally less than that of existing estimators. Furthermore, we provide Monte Carlo simulation results and an empirical data analysis that support our new method. Highlights • A new estimation method is proposed for generalized partially linear single-index models for longitudinal data. • The method is shown to be semiparametrically efficient. • Empirical results support the new method. [ABSTRACT FROM AUTHOR]

Abstract The composite quantile estimator is a robust and efficient alternative to the least-squares estimator in linear models. However, it is computationally demanding when the number of quantiles is large. We consider a model-averaged quantile estimator as a computationally cheaper alternative. We derive its asymptotic properties in high-dimensional linear models and compare its performance to the composite quantile estimator in both low- and high-dimensional settings. We also assess the effect on efficiency of using equal weights, theoretically optimal weights, and estimated optimal weights for combining the different quantiles. None of the estimators dominates in all settings under consideration, thus leaving room for both model-averaged and composite estimators, both with equal and estimated optimal weights in practice. Highlights • Comparing model averaged and composite quantile regression estimators. • Investigating properties of model averaged quantile estimators in high dimensions. • Studying the effect of the weight choice (estimated, optimal, equal weights). [ABSTRACT FROM AUTHOR]

Abstract This paper considers multidimensional jump type stochastic differential equations with super linear and non-Lipschitz coefficients. After establishing a sufficient condition for nonexplosion, this paper presents sufficient local non-Lipschitz conditions for pathwise uniqueness. The non-confluence property for solutions is investigated. Feller and strong Feller properties under local non-Lipschitz conditions are investigated via the coupling method. Sufficient conditions for irreducibility and exponential ergodicity are derived. As applications, this paper also studies multidimensional stochastic differential equations driven by Lévy processes and presents a Feynman–Kac formula for Lévy type operators. [ABSTRACT FROM AUTHOR]

Speech communication relies on articulatory and acoustic codes shared between speakers and listeners despite inter-individual differences in morphology and idiosyncratic articulatory strategies. This study addresses the long-standing problem of characterizing and modelling speaker-independent articulatory strategies and inter-speaker articulatory variability. It explores a multi-speaker modelling approach based on two levels: statistically-based linear articulatory models, which capture the speaker-specific articulatory variability on the one hand, are in turn controlled by a speaker model, which captures the inter-speaker variability on the other hand. A low dimensionality speaker model is obtained by taking advantage of the inter-speaker correlations between morphology and strategy. To validate this approach, contours of the vocal tract articulators were manually segmented on midsagittal MRI data recorded from 11 French speakers uttering 62 vowels and consonants. Using these contours, multi-speaker models with 14 articulatory components and two morphology and strategy components led to overall variance explanations of 66%–69% and root-mean-square errors of 0.36–0.38 cm obtained in leave-one-out procedure over the speakers. Results suggest that inter-speaker variability is more related to the morphology than to the idiosyncratic strategies and illustrate the adaptation of the articulatory components to the morphology. [ABSTRACT FROM AUTHOR]

Abstract When compared with Cartesian coordinates, curvilinear coordinates have shown significant superior accuracy when applied to a linear model for relative motion in a circular reference orbit. Experts assume the same superiority for an elliptical reference orbit. However, the curvilinear model in an elliptical reference orbit has not been established, and the superiority has not been proved. After a strictly theoretical derivation this paper establishes a linear model with spherical coordinates for relative motion in an elliptical reference orbit. The differential equation has the same form as the Lawden or Tschauner-Hempel equation and can be easily solved with the solutions from the literature. In addition both the theoretical derivation and the numerical examples prove the model's superior accuracy in leader-follower formation and formations with a large along-track distance. The proposed model reduces the approximate limits of close relative motion, and it is expected to be applied in the propagation, guidance, and control for longer term relative motion with large along-track distances. [ABSTRACT FROM AUTHOR]

Highlights • The elements squared of array covariance are exploited for non-coherent DOA estimation. • The nonlinear model is approximately cast as a linear model under a high-power reference source. • A simple but effective strategy is designed to avoid ambiguities. • The nested CADiS is adopted to enhance the number of DOFs greatly. • The algorithm performs independent of phase errors and can provide improved performance. Abstract This fast communication presents a non-coherent direction of arrival (DOA) estimation algorithm utilizing the elements squared of the array covariance. Very different from previous non-coherent DOA estimation algorithms, the nonlinear model produced by the squared operation is approximately cast as a linear model, under a high-power reference source. And then, the DOA estimation is efficiently obtained via the mature ℓ 1 -norm minimization. The proposed algorithm can be applied to arbitrary arrays. Focused on the estimation accuracy, the nested coprime array with displaced subarrays (CADiS) is exploited. To resolve the ambiguity problems in the scenario that the sources impinging on the array from the range [ − 90 ∘ , 90 ∘ ] , a simple but effective strategy is introduced. The proposed algorithm performs independent of the phase errors and can provide improved DOA estimation accuracy with reduced computational complexity in large snapshot case. Simulation results demonstrate the effectiveness and superiority of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Background: The 3rd Regional Comparative and Explanatory Study reports, analyses and compares academic results in mathematics, sciences, and reading for 15 Latin American countries. Validity is the foundation of a testing procedure, and the process of validation is important to the overall success of educational assessment as a whole. This methodological study deals specifically with an ecological point of view which includes and situates the person, process, context, and time of the testing situation. These descriptions indicate specific events showing how and what variables at the individual, school, or country level can give a deep understanding of the response process. The aims were to investigate ecological explanations of differential item functioning. Method: The study analysed the science test given in 2013 to 6th grade students and the data pool consisted of 12,657 students from 2,609 schools and 15 countries. A progressive inclusion of the variance distribution in different Bernoulli logistic regression models was carried out. Results: The analyses showed the presence of differential item functioning in 32% of the science items. Conclusions: The main source of differential item functioning was related to the human development level of the participating countries. [ABSTRACT FROM AUTHOR]

Abstract Introgressive crossbreeding between Red-Stirling and Chitralada Nile tilapia was carried out to assess the association of polymorphisms of short tandem repeats loci located in the growth hormone-I promoter and first intron with growth performance. The growth of seven genetic groups was assessed using a mixed linear model. We considered sex, age, and initial weight class of each animal as fixed effects, while genetic group, and polymorphisms in the short tandem repeats in the promoter and in the intron were considered as random effects. We found six alleles in the promoter, which were arranged in 18 of the 21 genotypic possible combinations. Alleles 191 and 196 of the promoter correspond to 80% of the assessed individuals' total alleles, and genotypes 196/196, 191/196 and 196/201 were the most frequent. We found only three alleles in the intron. Allelic variation at the promoter locus showed a significant association with weight, genotypes 181/181, 181/191 and 196/206 showed higher average weight, and genotypes 196/201, 201/201 and 191/196 showed the lowest weight. The polymorphism in the intron was not associated with fish weight. The approach utilized is unprecedented for short tandem repeats found in the growth hormone-I gene. The observed association indicates that this gene's polymorphism could be used as a tool in tilapia breeding programs. Highlights • Oreochromis niloticus is currently one of the most important aquaculture species worldwide. • Growth hormone plays an important role in post-natal somatic growth processes. • Microsatellite variation was detected in a GH promoter and intron in Nile tilapia. • Association of the GH promoter polymorphism with growth rate was demonstrated. [ABSTRACT FROM AUTHOR]

Abstract Recently, Vibrio harveyi has emerged as an important pathogenic disease with a high mortality rate (50%–70%) in Chinese tongue sole. The aim of this study was to estimate genetic parameters for resistance to V. harveyi in Chinese tongue sole and to evaluate the genetic correlation between resistance and growth traits (i.e. body weight and total length) (involving 50 full-sib families and 8547 individuals). Longitudinal linear models and cross-sectional threshold models were fitted by using different trait definitions (binary and categorical). After a 14-day test, the overall challenge test survival was 57.86% (ranging from 9.30% to 94.30% in families). The heritabilities of survival were ranging from 0.11 to 0.28, estimates obtained by linear models were higher than threshold models. The genetic correlations between resistance (binary and categorical traits) and two growth traits were moderately positive (0.27–0.51). Very high Pearson and Spearman correlations (0.99 to 1) of full-sib family EBVs between different models were all close to unity which might reveal the similar predictive ability of the four models. The favourable heritabilities and moderate positive genetic correlations indicate that joint genetic improvement of vibriosis resistance and growth performance would be feasible. Statement of relevance This paper offers guidelines to disease-resistance selective breeding strategies in Chinese tongue sole. Highlights • Genetic variation for disease resistance traits were evaluated by challenge test with Vibrio harveyi in Chinese tongue sole. • Two types of models (cross-sectional and longitudinal models) were used for genetic analyses based on different trait definitions (i.e., binary test survival and categorical test day survival). • Low to moderate heritabilities for V. harveyi resistance traits were estimated, indicating that resistance could be genetically improved by selective breeding. • Genetic correlations between disease resistance traits and growth traits were moderately positive (ranging from 0.27 to 0.51) and significantly different from zero. • Estimated breeding values (EBVs) of the four models showed high and positive correlations (0.99–1), which meant that these models demonstrated similar utility for predicting breeding value. [ABSTRACT FROM AUTHOR]

Background: Several methods to handle data generated from bottom-up proteomics via liquid chromatography-mass spectrometry, particularly for peptide-centric quantification dealing with post-translational modification (PTM) analysis like reversible cysteine oxidation are evaluated. The paper proposes a pipeline based on the R programming language to analyze PTMs from peptide-centric label-free quantitative proteomics data. Results: Our methodology includes variance stabilization, normalization, and missing data imputation to account for the large dynamic range of PTM measurements. It also corrects biases from an enrichment protocol and reduces the random and systematic errors associated with label-free quantification. The performance of the methodology is tested by performing proteome-wide differential PTM quantitation using linear models analysis (limma). We objectively compare two imputation methods along with significance testing when using multiple-imputation for missing data. Conclusion: Identifying PTMs in large-scale datasets is a problem with distinct characteristics that require new methods for handling missing data imputation and differential proteome analysis. Linear models in combination with multiple-imputation could significantly outperform a t-test-based decision method. [ABSTRACT FROM AUTHOR]

Abstract In this paper, an order-reduction modeling method is proposed to capture the dynamic performance of a bolted joint structure. The model contains the influences of nonlinear contact force, damping and geometrical characteristics of connecting interface. The equations of motion of a bolted joint structure are divided into the linear characteristics and the nonlinear characteristics based on desired locations information. A new hybrid coordinate reduction (HCR) technique is introduced where the nonlinear characteristics within connecting interface are retained into the physical space while other linear modes are truncated. However, the error caused by truncating higher-order linear modes is compensated using a residual flexible matrix. The connecting interface of the bolted joint structure is condensed to a visual point via the coordinate transformation. The main advantage of this method is the reduced computational cost for solving the nonlinear dynamic equations. A series of experimental tests are performed to demonstrate the effectiveness of the proposed method. It is shown that the proposed method provides a more realistic tool for capturing the nonlinear dynamics and performance of a complex joint structure. [ABSTRACT FROM AUTHOR]

Estimating the timing of the occurrence of events that characterize growth cycles in vegetation from time series of remote sensing data is desirable for a wide area of applications. For example, the timings of plant life cycle events are very sensitive to weather conditions and are often used to assess the impacts of changes in weather and climate. Likewise, understanding crop phenology can have a large impact on agricultural strategies. To study phenology using remote sensing data, the timings of annual phenological events must be estimated from noisy time series that may have many missing values. Many current state-of-the-art methods consist of smoothing time series and estimating events as features of smoothed curves. A shortcoming of many of these methods is that they do not easily handle missing values and require imputation as a preprocessing step. In addition, while some currently used methods may be extendable to allow for temporal uncertainty quantification, uncertainty intervals are not usually provided with phenological event estimates. We propose methodology utilizing Bayesian dynamic linear models to estimate the timing of key phenological events from remote sensing data with uncertainty intervals. We illustrate the methodology on weekly vegetation index data from 2003 to 2007 over a region of southern India, focusing on estimating the timing of start of season and peak of greenness. Additionally, we present methods utilizing the Bayesian formulation and MCMC simulation of the model to estimate the probability that more than one growing season occurred in a given year. Supplementary materials accompanying this paper appear online. [ABSTRACT FROM AUTHOR]

Abstract In the present study we aimed at reaching a procedure for control of factors that have a potential to contribute to bioanalytical method bias and consistent dried blood spot (DBS) formation. The most influencing factors were identified via experimental design and image processing methodology. The effects of five qualitative factors - temperature of blood samples, type of pipettes, pipetting technique, age of blood samples and analysts were investigated by multilevel categorical D-optimal design. Due to well-known influence of hematocrit (Hct) level on rheological characteristic of blood samples, five responses were observed through the study (RSD 22%Hct , RSD 30%Hct , RSD 39%Hct , RSD 51%Hct , RSD 62%Hct) for their ability to provide insights into influence of factor settings onto DBS areas consistency. The area of blood spots was determined by image processing after scanning DBS cards with four spot replicates, corresponding to particular combination of investigated factors, as defined by experimental plan. The qualitative linear mathematical models with added appropriate two-factor interactions were derived by the principle of backward elimination. It was concluded that %RSD value of DBS, for all investigated Hct levels, is completely independent of type of pipettes and age of blood samples, but can significantly be affected by blood sample temperature, pipetting technique and training of analyst. Therefore, the procedure for precise and accurate formation of DBS of uniform area, regardless the Hct value, implies samples at body/room temperature, reversed pipetting technique for rigorous delivery of a sample volume onto the card and a properly trained analyst for handling samples. Finally, a verification experiment was performed and the adequacy of the suggested procedure was confirmed. Identification of the factors affecting the consistency of DBS formation provided the evidence that this contribution to total assay bias can be successfully controlled and reduced. Graphical abstract Unlabelled Image Highlights • The procedure for consistent DBS formation, regardless the sample Hct value, is proposed. • Multilevel categorical D-optimal design was used for identification of influential factors. • Image processing ensured straightforward and reliable estimation of scanned DBSs area. • %RSD value of DBSs is completely independent of type of pipettes and age of blood samples. • The source of variability is blood sample temperature, pipetting technique and training of analyst. [ABSTRACT FROM AUTHOR]

Highlights • Widely linear filter model is used to represent temporally correlated source signal. • A new blind source separation algorithm is proposed combining widely linear filter model identification with joint matrix diagonalization. • A special joint matrix diagonalization algorithm is proposed to adapt to the proposed blind source separation algorithm. Abstract In this paper, we introduce a new blind source separation (BSS) method for temporal correlated noncircular sources that uses widely linear filter (WLF) model to efficiently describe the temporal structure. The algorithm consists of a WLF coefficients estimator followed by complex matrix joint diagonalization. In the derivation of the new BSS algorithm, an emerging matrix joint diagonalization issue needs to be addressed. Subsequently, a complex matrix joint diagonalization algorithm based on successive Shear and Givens rotations (SGR) is proposed to deal with the new joint diagonalization problem. Simulations of SGR show that it converges fast and converges to excellent value, and it shows a satisfactory anti-perturbation performance of matrix for the new BSS algorithm. We compare the performance of the new BSS algorithm with several competing algorithms, and the simulation results demonstrate the superior performance of the proposed BSS algorithm for both noncircular complex Gaussian and non-Gaussian sources. [ABSTRACT FROM AUTHOR]

In this paper, an efficient indoor localization algorithm based on the confidence-interval fuzzy model is presented. The width of the confidence interval is essential within the proposed fingerprinting method for calculating weights, which are then taken into account while searching for the $K$ nearest neighbors in the database of fingerprints. For each beacon in the test room, a new confidence-interval fuzzy path-loss model composed of several local linear models is constructed. The map of fingerprints is then constructed of a set of confidence-interval fuzzy models. By their consideration, the localization accuracy is significantly improved in comparison with other commonly used path-loss models. The most important novelty of this paper is the introduction of the confidence interval within the fingerprinting method, which additionally improves localization results. The platform of the localization system is developed on the basis of a smartphone and Bluetooth beacons. Therefore, the localization algorithm has to be optimized in order to be computationally efficient, which is essential for real-time processing and low energy consumption on a smartphone. [ABSTRACT FROM AUTHOR]

In this paper, we consider the mixture of sparse linear regressions model. Let $\boldsymbol {\beta }^{(1)},\ldots, \boldsymbol {\beta }^{(L)}\in \mathbb {C} ^{n}$ be $L $ unknown sparse parameter vectors with a total of $K $ non-zero elements. Noisy linear measurements are obtained in the form $y_{i} = \boldsymbol {x}_{i} ^{\mathrm{ H}} \boldsymbol {\beta }^{(\ell _{i})} + w_{i}$ , each of which is generated randomly from one of the sparse vectors with the label $\ell _{i} $ unknown. The goal is to estimate the parameter vectors efficiently with low sample and computational costs. This problem presents significant challenges as one needs to simultaneously solve the demixing problem of recovering the labels $\ell _{i} $ as well as the estimation problem of recovering the sparse vectors $\boldsymbol {\beta }^{(\ell)} $. Our solution to the problem leverages the connection between modern coding theory and statistical inference. We introduce a new algorithm, Mixed-Coloring, which samples the mixture strategically using query vectors $\boldsymbol {x}_{i} $ constructed based on ideas from sparse graph codes. Our novel code design allows for both efficient demixing and parameter estimation. To find $K$ non-zero elements, it is clear that we need at least $\Theta (K)$ measurements, and thus the time complexity is at least $\Theta (K)$. In the noiseless setting, for a constant number of sparse parameter vectors, our algorithm achieves the order-optimal sample and time complexities of $\Theta (K)$. In the presence of Gaussian noise, for the problem with two parameter vectors (i.e., $L = 2$), we show that the Robust Mixed-Coloring algorithm achieves near-optimal $\Theta (K \mathop {\mathrm {polylog}}\nolimits (n))$ sample and time complexities. When $K = \mathcal {O}(n^{\alpha })$ for some constant $\alpha \in (0,1)$ (i.e., $K$ is sublinear in $n$), we can achieve sample and time complexities both sublinear in the ambient dimension. In one of our experiments, to recover a mixture of two regressions with dimension $n = 500$ and sparsity $K = 50$ , our algorithm is more than 300 times faster than EM algorithm, with about one third of its sample cost. 1 The proposed algorithm works even when the noise is non-Gaussian in nature, but the guarantees on sample and time complexities are difficult to obtain. [ABSTRACT FROM AUTHOR]