Variance-covariance Research Articles

Evolutionary and Ecological Processes Determining the Properties of the G Matrix.

AbstractThe G matrix is the matrix of additive genetic variances and covariances for a vector of phenotypes. Here we apply the classical theory for the balance among selection, genetic drift, and mutations to find the contributions to G from each locus for populations at stasis. The fitness is approximated by a linear function of phenotypes, with coefficients affected by environmental fluctuations. We show that the G matrix can be decomposed into four additive components generated by selection, drift, mutations, and environmental fluctuations. Selection is on average counteracted by the other three processes included in Fisher's concept of deterioration of the environment, generating considerable changes in mean phenotypes. The theory illustrates that neither Fisher's fundamental theorem nor Lande's classical gradient formula is sufficient for assessing adaptive changes through time unless the deteriorations are corrected for. This applies for populations at stasis, but also for populations that are subject to long-term evolutionary changes. The theory also indicates several possible comparative studies for investigations of deteriorating effects. Our analyses also suggest that the factor loadings to the eigenvector of the G matrix with the lowest eigenvalue will rather accurately indicate the relative contributions from different phenotype components to fitness. This is information notoriously difficult to obtain in natural populations.

Academic procrastination and inappropriate use of social networks among university students

ABSTRACT Academic procrastination is the tendency to delay academic activities such as carrying out tasks, studying for exams and similar. This behaviour may be related to the inappropriate use of social networks, especially by those students who procrastinate the most. The aim of this study is to identify latent profiles of academic procrastination and determine whether there are differences in terms of the areas of delay (academic activities, final degree projects, studying for exams and keeping abreast of academic tasks) and the factors of inadequate use of social networks (obsession with networks, a lack of better control and excessive use). The participants in this study were 1784 students from 24 Spanish universities. The Spanish version of the procrastination assessment scale and the questionnaire on addiction to social networks were used for information gathering. A Latent Profile Analysis, multivariate analysis of variance (MANOVA) and multivariate covariance analysis (MANCOVA) were carried out to examine the differences between the profiles detected, delay areas and social network addiction factors. The results showed three latent profiles in procrastination behaviour: low, moderate and high procrastination. Moreover, significant differences were observed in terms of the areas of delay and network addiction factors. In conclusion, the greater the procrastinating behaviour, the greater the tendency to delay academic activities and present more frequently inadequate conduct in the use of social networks.

Effects Of Jolly Phonics Strategy on Pupils’ Reading Achievement and Attitudes in Lagos State Lower Basic Schools

This study investigated the effectiveness of implementing the Jolly Phonics strategy on pupils’ reading achievement and attitude within the context of lower basic schools in Lagos State, Nigeria. Two research questions and two hypotheses guided the study. The study adopted the quasi-experimental pre-testing, post-test, control group experimental design. The population of the study consisted of all basic 1 pupil in public lower basic schools in Lagos State. While the sample size was Three hundred and twenty-four (324) pupils of Lower Basic 1 and Nine (9) teachers who were randomly and purposively selected using Simple Random sampling technique. The sample size was drawn from three schools from three Local Government Education Authority Areas (LGEAs) in Lagos State public lower basic schools. Two research questions were used and three instruments used for eliciting responses were Achievement Test in Reading (ATR), Pupils’ Attitude to Reading Questionnaire (PARQ) and Pupils’ Verbal Ability Test (PVAT). The instruments were validated by experts in early childhood and measurement and evaluation. The test re-test method was adopted for instrument reliability at 0.75. The data collected were analyzed using descriptive statistics (Mean and Standard deviation) Variance and Analysis Covariance (ANCOVA) Findings revealed that Jolly Phonics had significant effect on the academic achievement in reading of basic 1 pupil in lower basic schools and their attitude. Pupils taught with Jolly Phonics strategy performed better in reading achievement and had positive attitude than those taught with conventional method. It was recommended that Jolly phonics instructional strategies should be incorporated into basic school reading curriculum. This will help them develop reading skills easily. Individual differences in pupils’ reading ability should be identified and addressed through variation of reading strategies as was provided for in the experimental groups with jolly phonics strategy. Government, school managers and administrators should provide adequate instructional materials that will motivate pupils to easily adapt to jolly instructional strategy.

Structured prior distributions for the covariance matrix in latent factor models

Factor models are widely used for dimension reduction in the analysis of multivariate data. This is achieved through decomposition of a p×p\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$p \ imes p$$\\end{document} covariance matrix into the sum of two components. Through a latent factor representation, they can be interpreted as a diagonal matrix of idiosyncratic variances and a shared variation matrix, that is, the product of a p×k\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$p \ imes k$$\\end{document} factor loadings matrix and its transpose. If k≪p\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$k \\ll p$$\\end{document}, this defines a parsimonious factorisation of the covariance matrix. Historically, little attention has been paid to incorporating prior information in Bayesian analyses using factor models where, at best, the prior for the factor loadings is order invariant. In this work, a class of structured priors is developed that can encode ideas of dependence structure about the shared variation matrix. The construction allows data-informed shrinkage towards sensible parametric structures while also facilitating inference over the number of factors. Using an unconstrained reparameterisation of stationary vector autoregressions, the methodology is extended to stationary dynamic factor models. For computational inference, parameter-expanded Markov chain Monte Carlo samplers are proposed, including an efficient adaptive Gibbs sampler. Two substantive applications showcase the scope of the methodology and its inferential benefits.

Low frequency load identification under high noise level using weighted total least squares

Load identification in structural dynamics is an ill-conditioned inverse problem, and the errors exist in both the transfer function matrix and the response vector, which have a great influence on the accuracy of the identification, especially at the low frequency and high noise. Although the Total Least Squares (TLS) method can consider the errors in the transfer function matrix and the response vector, the problem of unequal precision between them has not been properly solved. In this article, the Weighted Total Least Squares (WTLS) method for load identification is presented. First, the variance–covariance matrices are constructed using stochastic information, which can consider the unequal precision of the data and make the solution less biased and more precise. Then, Tikhonov regularization is used to regularize the WTLS method to reduce the ill-condition of the transfer function matrix. To validate the performance of the WTLS method under different noise levels, a load identification simulation with four excitation loads on a plate and a load identification experiment with three excitation loads on a steel plate were studied. The results show that when the noise levels of the transfer function and the response are different, the accuracy of the TLS solution is poor. Especially when the noise levels are high, the identification deviations are large. The WTLS method considers the unequal noise levels using the weight matrices, which provides more stable load estimates at different noise levels. And the identification deviations are greatly reduced compared to the TLS method.

Comparative Analysis of Value-at-Risk in Market Risk Prediction in Banks Using GARCH Volatility

This study aims to compare the disclosure of Value at Risk (VaR) in market risk prediction among banks in Indonesia. By employing comparative and analytical methods, this research examines the effectiveness of VaR disclosure as a market risk prediction tool. Through the evaluation of VaR models disclosed by Indonesian banks and their comparison to a parametric model using asymmetric GARCH volatility for Variance Covariance Value at Risk, this study identifies the extent to which VaR disclosure can be relied upon to predict market risk. This research contributes to the understanding of risk management practices in the Indonesian banking sector and offers recommendations for improving market risk prediction accuracy through more effective VaR disclosures.

Forecasting interval‐valued returns of crude oil: A novel kernel‐based approach

AbstractThis paper proposes a novel kernel‐based generalized random interval multilayer perceptron (KG‐iMLP) method for predicting high‐volatility interval‐valued returns of crude oil. The KG‐iMLP model is constructed by utilizing the distance based on a kernel function, which outperforms the conventional Euclidean distance. Additionally, the optimal kernel function is estimated using the variance–covariance matrix of the prediction error, contributing to a better understanding of the overall characteristics of interval‐valued data. The introduction of the kernel function renders the algorithms used for estimating machine learning parameters ineffective. Therefore, this paper further proposes a backward distance of accumulative error propagation algorithm to estimate both the kernel function and model parameters, which provides a feasible approach for utilizing kernel function in interval neural networks. In the empirical analysis of weekly and daily returns of WTI crude oil, the superior predictive performance of the proposed method is demonstrated, enabling stable and accurate predictions for both point values and interval values. The model exhibits consistent outstanding performance across different network structures, showcasing the potential of KG‐iMLP for crude oil price forecasting.

Elevating performance in BDS-3 Precise Point Positioning Ambiguity Resolution through LEO augmented constellations

With the help of constellations combined BDS-3 (BeiDou Navigation Satellite System) and LEO (Low Earth Orbit), we aim to improve the performances of PPP (Precise Point Positioning) and PPP Ambiguity Resolution (PPP-AR). First, LEO constellations with 66, 96, 156, 192, 270, and 288 satellites are designed to establish six LEO augmented constellations. Based on four challenge scenarios, PPP of six LEO augmented constellations are performed. Even in the strip-shaped obstruction area losing more than 60% of observations, the positioning accuracies of static PPP are improved by (1.0%, 5.7%, 0.2%) for BDS+66LEO, and (14.9%, 33.7%, 7.5%) for BDS+288LEO in the east, north, and up components. The positioning accuracy and convergence time improve quickly as the number of LEO satellites increased from 66 to 156, and slowdown from 192 to 288. Two constellations, BDS+156LEO and BDS+192LEO, are considered to strike the balance of acceptable PPP performance and constellation load. Then, the performance of PPP-AR is realized with the estimated UPD (Uncalibrated Phase Delay). The correlations among geodetic parameters and ambiguities deriving from the post-fit variance–covariance matrix are analyzed to illustrate the enhancement of LEO augmented constellations. Under the condition of 20°elevation angle occlusion with azimuth ranging from 0 to 360°, BDS+156LEO can achieve an ambiguity fix rate of up to 98.37% with a faster mean TTFF (Time To First Fix) of 12.76 min, and the positioning accuracies in the east, north, and up components are improved by 28.2%, 21.4%, and 15.1% respectively, compared with BDS-3. BDS+192LEO realized an 98.42% ambiguity fixing rate, and improved BDS-3 by 31.4%, 26.7% and 11.5% with its TTFF been 12.25 min.

Entropy associated with conformational and solvent-density fluctuations in biomolecular solutions

Microscopic formula to describe the entropy of biomolecular solutions are derived based on the Gibbs formula of entropy, and the generalized Langevin theory combined with the RISM/3D-RISM theory. Two formula are derived: one is concerned with the conformational fluctuation of a biomolecule, and the other with the density fluctuation of solvent around a solute. The formula derived for the entropy associated with the conformational fluctuation is Sconf=12kBlog2π3NA+3N2kBwhere N is the number of atoms in the solute, and A is the determinant of the inverse of the variance–covariance matrix of conformational fluctuation.The formula for the entropy of solvent at a pair of positions around a solute is also derived to be, ssolv(r,r′)=12kBlog2πnB+3n2kBwhere n is the number of atoms in a solvent molecule, and B is essentially the determinant of the matrix of the density-pair-correlation functions. The entropy at a local position r may be obtained by integrating the expression by r′ over the entire volume of the system.The feasibility of the calculation to find the entropies is discussed.

Determinants of regional business cycle synchronization in Greece

AbstractWe assess the determinants of regional business cycle synchronization in Greece vis‐à‐vis the national reference business cycle, using NUTSII annual data. The computation of the time‐varying synchronization is based on the dynamic estimate of a conditional variance–covariance model and subsequently, a panel regression model is used to evaluate its determinants. The findings show that island regions, industrial structure, imports, savings and disposable income are the key determinants, based on the GVA business cycle synchronization vis‐à‐vis the national reference cycle. We also assess the determinants of employment synchronization (vis‐à‐vis the national employment level) and we find that regions with higher disposable income and public spending tend to drive the level of synchronization. Turning to the inter‐regional synchronization, we provide evidence that investments, disposable income and employment drive the GVA business cycle synchronization, whereas the employment synchronization is determined by the level of imports, disposable income and public spending, as well as by the status of regions as island economies. We further show that the Greek economic crisis during the period 2010–2018 has diminished or eliminated the effects of the aforementioned drivers, suggesting that during the said period, synchronization was mainly driven by the wider economic conditions. These findings lead to important policy implications, which are thoroughly discussed.

Interspecific selection in a diverse mycorrhizal symbiosis.

Coevolution describes evolutionary change in which two or more interacting species reciprocally drive each other's evolution, potentially resulting in trait diversification and ecological speciation. Much progress has been made in analysis of its dynamics and consequences, but relatively little is understood about how coevolution works in multispecies interactions, i.e., those with diverse suites of species on one or both sides of an interaction. Interactions among plant hosts and their mutualistic ectomycorrhizal fungi (ECM) may provide an ecologically unique arena to examine the nature of selection in multispecies interactions. Using native genotypes of Monterey pine (Pinus radiata), we performed a common garden experiment at a field site that contains native stands to investigate selection from ECM fungi on pine traits. We planted seedlings from all five native populations, as well as inter-population crosses to represent intermediate phenotypes/genotypes, and measured seedling traits and ECM fungal traits to evaluate the potential for evolution in the symbiosis. We then combined field estimates of selection gradients with estimates of heritability and genetic variance-covariance matrices for multiple traits of the mutualism to determine which fungal traits drive plant fitness variation. We found evidence that certain fungal operational taxonomic units, families and species-level morphological traits by which ECM fungi acquire and transport nutrients exert selection on plant traits related to growth and allocation patterns. This work represents the first field-based, community-level study measuring multispecific coevolutionary selection in nutritional symbioses.

#782 Latent profiles of patients undergoing maintenance hemodialysis based on hemodynamic indicators

Abstract Background and Aims Both intradialytic hypotension and intradialytic hypertension are conditions occurring frequently during hemodialysis and are associated with unfavorable clinical outcomes. However, they remain incompletely understood and poorly defined in the literature. Since intradialytic parameters are now captured automatically in electronic medical records in many institutions, machine learning could emerge as an innovative tool for hemodialysis clinical research. Latent Profile Analysis is a model-based clustering method. It allows to identify hidden subpopulations derived from a set of observed indicator variables, maximizing between-group differences and within-group similarities. Method The objectives of this study were to identify latent profiles using a set of hand-crafted indicators and to interpret these subgroups of patients in a clinically meaningful manner. We conducted a retrospective single-center cohort study of adult patients receiving hemodialysis between January 2017 and December 2022 in Montreal. Given the highly heterogeneous nature of this population, we only enrolled incidental patients. The data were extracted from the NephroCare database, which is the clinical information system used by dialysis teams in our center. Sixteen indicators were derived from time series of blood pressure and heart rate measurements, capturing various patient-centric aspects of intradialytic hemodynamics, including trends, zeniths, nadirs, and time-related features. Latent Profile Analysis involved fitting a series of models, and the best model was selected based on fit indices, evaluation metrics, and clinical interpretation of the distribution of indicator within identified profiles. The primary analysis used complete-case data, and a sensitivity analysis assessed the impact of missing data through multiple imputation. Internal model validation was conducted using k-fold cross-validation. Results The selected model consisted of four profiles with varying variance and null covariance across indicators. Fig. 1 presents radar plots summarizing the identified profiles. Profile 1 included patients with prolonged early nadirs and a nadir-to-zenith transition pattern at the session level. Profile 2 was characterized by frequent nadirs and zeniths without a specific timing preference. Profile 3 comprised patients with infrequent blood pressure variations altogether. Profile 4 included patients with frequent early nadirs but no zeniths. The model demonstrated good generalization to unseen cases and exhibited relative robustness to missing data. Conclusion In this preliminary study, we demonstrated that Latent Profile Analysis can identify clinically meaningful subpopulations of hemodialysis patients based on hemodynamic indicators. The association between the identified profiles and hard clinical endpoints remains to be established.

COMPARISON MRCD AND ORACLE FOR ESTIMATING THE DETERMINANT OF HIGH DIMENSIONAL COVARIANCE MATRIX

Estimating the variance matrix has an important role in statistical applications and conclusions, in high–dimensional matrices if the number of variables is greater than the number of observations P > n, the traditional statistical methods are not reliable because they give uncontrolled estimates. Shrinkage methods are used to estimate the high–dimensional variance matrix. In this research, the high–dimensional variance matrix was estimated using the robust Nonparametric method Minimum Regularized Covariance Determinant (MRCD), which is based on Mahalanobis distance, and compared with the variance matrix estimated by the Oracle method, which is based on the Frobenius criterion

A cautionary tale on the effects of different covariance structures in linear mixed effects modeling of fMRI data.

With the steadily increasing abundance of longitudinal neuroimaging studies with large sample sizes and multiple repeated measures, questions arise regarding the appropriate modeling of variance and covariance. The current study examined the influence of standard classes of variance-covariance structures in linear mixed effects (LME) modeling of fMRI data from patients with pediatric mild traumatic brain injury (pmTBI; N = 181) and healthy controls (N = 162). During two visits, participants performed a cognitive control fMRI paradigm that compared congruent and incongruent stimuli. The hemodynamic response function was parsed into peak and late peak phases. Data were analyzed with a 4-way (GROUP×VISIT×CONGRUENCY×PHASE) LME using AFNI's 3dLME and compound symmetry (CS), autoregressive process of order 1 (AR1), and unstructured (UN) variance-covariance matrices. Voxel-wise results dramatically varied both within the cognitive control network (UN>CS for CONGRUENCY effect) and broader brain regions (CS>UN for GROUP:VISIT) depending on the variance-covariance matrix that was selected. Additional testing indicated that both model fit and estimated standard error were superior for the UN matrix, likely as a result of the modeling of individual terms. In summary, current findings suggest that the interpretation of results from complex designs is highly dependent on the selection of the variance-covariance structure using LME modeling.

Advances in meta-analysis: A unifying modelling framework with measurement error correction.

In psychological studies, multivariate outcomes measured on the same individuals are often encountered. Effects originating from these outcomes are consequently dependent. Multivariate meta-analysis examines the relationships of multivariate outcomes by estimating the mean effects and their variance-covariance matrices from series of primary studies. In this paper we discuss a unified modelling framework for multivariate meta-analysis that also incorporates measurement error corrections. We focus on two types of effect sizes, standardized mean differences (d) and correlations (r), that are common in psychological studies. Using generalized least squares estimation, we outline estimated mean vectors and variance-covariance matrices for d and r that are corrected for measurement error. Given the burgeoning research involving multivariate outcomes, and the largely overlooked ramifications of measurement error, we advocate addressing measurement error while conducting multivariate meta-analysis to enhance the replicability of psychological research.

Equity markets and ESG dynamics: Assessing spillovers and portfolio strategies through time-varying parameters

Motivated by the recent surge of socially responsible activism, we examine the return and volatility transmission between the global Environmental, Governance, and Social index and traditional equity markets of G7 countries. For this purpose, we employ the TVP-VAR model on the returns and volatilities from 04 April 2011 to 09 September 2023. The variance co-variance matrix of the TVP-VAR model is used for portfolio analysis. Our finding indicates a significant level of connectedness between green and traditional equity markets. Moreover, green equity functions as a transmitter of both return and volatility spillovers to the traditional equity markets. In addition, our analysis of dynamic connectedness reflects a spike in both total return and volatility spillover amid the COVID-19 period. This indicates that the relationship between green and traditional equity markets intensifies during times of market stress and uncertainty which further infer implications for portfolio diversification. Furthermore, we calculated optimal asset allocation within portfolios and identified effective hedging ratios for different ESG and G7 equity pairs. The implications of the study include informing portfolio diversification strategies, and guiding decision-making for policymakers, hedge fund managers, and portfolio professionals amidst market uncertainties, based on identified market dynamics and connectedness.

Low-pressure measurements of energy depositions in nanometre objects with a single ionisation chamber in a 60Co beam using the variance–covariance method

Dose-mean lineal energies were measured using the variance–covariance method and commercial ionisation chambers in a therapy level 60Co field at several different air pressures. The results confirmed the trend from previous studies and the experimental data was extended down to a simulated object diameter of 4.2 nm using a low-noise electrometer. The variance–covariance method was adapted to consecutive charge integrations from a single ionisation chamber and used to correct for signal variations due to pressure drifts in the ion chamber. Monte Carlo simulations using MCNP 6.2 were performed, and the results follow the trend of the experimental values in the nanometric region.

Three-dimensional biphase fabric estimation from 2D images by deep learning

A pruned VGG19 model subjected to Axial Coronal Sagittal (ACS) convolutions and a custom VGG16 model are benchmarked to predict 3D fabric descriptors from a set of 2D images. The data used for training and testing are extracted from a set of 600 3D biphase microstructures created numerically. Fabric descriptors calculated from the 3D microstructures constitute the ground truth, while the input data are obtained by slicing the 3D microstructures in each direction of space at regular intervals. The computational cost to train the custom ACS-VGG19 model increases linearly with p (the number of images extracted in each direction of space), and increasing p does not improve the performance of the model - or only does so marginally. The best performing ACS-VGG19 model provides a MAPE of 2 to 5% for the means of aggregate size, aspect ratios and solidity, but cannot be used to estimate orientations. The custom VGG16 yields a MAPE of 2% or less for the means of aggregate size, distance to nearest neighbor, aspect ratios and solidity. The MAPE is less than 3% for the mean roundness, and in the range of 5-7% for the aggregate volume fraction and the mean diagonal components of the orientation matrix. Increasing p improves the performance of the custom VGG16 model, but becomes cost ineffective beyond 3 images per direction. For both models, the aggregate volume fraction is predicted with less accuracy than higher order descriptors, which is attributed to the bias given by the loss function towards highly-correlated descriptors. Both models perform better to predict means than standard deviations, which are noisy quantities. The custom VGG16 model performs better than the pruned version of the ACS-VGG19 model, likely because it contains 3 times (p = 1) to 28 times (p = 10) less parameters than the ACS-VGG19 model, allowing better and faster cnvergence, with less data. The custom VGG16 model predicts the second and third invariants of the orientation matrix with a MAPE of 2.8% and 8.9%, respectively, which suggests that the model can predict orientation descriptors regardless of the orientation of the input images.

Evolutionary potential under heat and drought stress at the southern range edge of North American Arabidopsis lyrata.

The warm edges of species' distributions are vulnerable to global warming. Evidence is the recent range retraction from there found in many species. It is unclear why populations cannot easily adapt to warmer, drier, or combined hot and dry conditions and locally persist. Here, we assessed the ability to adapt to these stressors in the temperate species Arabidopsis lyrata. We grew plants from replicate seed families of a central population with high genetic diversity under a temperature and precipitation regime typical of the low-latitude margin or under hotter and/or drier conditions within naturally occurring amplitudes. We then estimated genetic variance-covariance (G-) matrices of traits depicting growth and allocation as well as selection vectors to compare the predicted adaptation potential under the different climate-stress regimes. We found that the sum of genetic variances and genetic correlations were not significantly different under stress as compared to benign conditions. However, under drought and heat drought, the predicted ability to adapt was severely constrained due to strong selection and selection pointing in a direction with less multivariate genetic variation. The much-reduced ability to adapt to dry and hot-dry conditions is likely to reduce the persistence of populations at the low-latitude margin of the species' distribution and contribute to the local extinction of the species under further warming.

Estimating time-varying factors’ variance in the string-term structure model with stochastic volatility

This paper presents a novel term structure of interest rate (TSIR) model with stochastic volatility and jumps (SVJ) that combines the market framework proposed by Brace et al. (1997) with the string-shock framework of Santa-Clara and Sornette (2001). In this model, the factors’ variance is estimated through the eigendecomposition of a variance–covariance matrix obtained with a measure of market volatility derived from out-of-the-money option prices and historical correlations of interest rates traded in the futures market. The stochastic evolution of the factors’ variance is governed by the 4/2 model developed by Grasselli (2017), including jumps. A novel method is employed to estimate the parameters of the SVJ model that minimizes the distance between the sample moments and the moments of a gamma distribution. The empirical application of the model in the Brazilian derivatives market demonstrates its effectiveness in accurately capturing the volatility smile of interest rate options.