Variance Of Coefficients Research Articles

Evaluation for Parkinsonian Bradykinesia by deep learning modeling of kinematic parameters.

A wearable sensor system is available for monitoring of bradykinesia in patients with Parkinson's disease (PD), however, it remains unclear whether kinematic parameters would reflect clinical severity of PD, or would help clinical diagnosis of physicians. The present study investigated whether the classification model using kinematic parameters from the wearable sensor may show accordance with clinical rating and diagnosis in PD patients. Using the Inertial Measurement Units (IMU) sensor, we measured the movement of finger tapping (FT), hand movements (HM), and rapid alternating movements (RA) in 25 PD patients and 21 healthy controls. Through the analysis of the measured signal, 11 objective features were derived. In addition, a clinician who specializes in movement disorders viewed the test video and evaluated each of the Unified Parkinson's Disease Rating Scale (UPDRS) scores. In all items of FT, HM, RA, the correlation between the linear regression score obtained through objective features (angle, period, coefficient variances for angle and period, change rates of angle and period, angular velocity, total angle, frequency, magnitude, and frequency × magnitude) and the clinician's UPDRS score was analyzed, and there was a significant correlation (rho > 0.7, p < 0.001). PD patients and controls were classified by deep learning using objective features. As a result, it showed a high performance with an area under the curve (AUC) about as high as 0.9 (FT Total = 0.950, HM Total = 0.889, RA Total = 0.888, ALL Total = 0.926. This showed similar performance to the classification result of binary logistic regression and neurologist, and significantly higher than that of family medicine specialists. Our results suggest that the deep learning model using objective features from the IMU sensor can be usefully used to identify and evaluate bradykinesia, especially for general physicians not specializing in neurology.

Enhancement of mesoscopic fluctuations in transmission of light through a disordered medium at grazing angles

We study how intensity fluctuations in transmission through a disordered slab change depending on the regime of wave transport. A system with large (compared to the wavelength of light) inhomogeneities is considered. Within a diagrammatic approach, the variance of the total transmission coefficient is calculated numerically beyond the diffusion approximation. A great enhancement of fluctuations is found in the crossover from the ballistic to the diffusive transport at the grazing angles of incidence on the sample surface. The effect originates from the reflection of waves that propagate nearly parallel to the sample boundaries and experience scattering through small angles, and reveals itself both in transmission and reflection and in correlations between the transmitted and reflected fluxes.

Detectability-Based JPEG Steganography Modeling the Processing Pipeline: The Noise-Content Trade-off

The current art of steganography shows that schemes using a deflection criterion (such as MiPOD) for JPEG steganography are usually subpar with respect to distortion-based schemes. We link this lack of performance to a poor estimation of the variance of the model of the noise on the cover image. However, this statistically-based method provides a better assessment of the detectability of hidden data as well as theoretical guarantees under a given model. In this paper, we propose a method to obtain better estimates of the variances of DCT coefficients by taking into account the dependencies introduced by development pipeline on pixels. A second method, which is a side-informed extension of Gaussian Embedding in the JPEG domain using quantization error as side-information, is also formulated and shown to achieve state-of-the-art performances. Eventually, the trade-off between noise and content complexity in steganography is thoroughly analyzed through the lenses of these two new methods using a wide range of numerical experiments.

Study of genetic variability in relation to yield and yield components in forage sorghum (Sorghum bicolor L. Moench)

The present investigation was carried out the genetic analysis of breeding material through components of variance, nature and magnitude of gene effects, genetic variability, heritability, genetic advance, correlation coefficients, general and specific combining ability effects and heterosis among yield and it components in parents and F1’s hybrids. Ten diverse parents i.e., UP Chari-1, HC-308, UP Chari-2, HC-171, Pant Chari-8, CSV-17, Pant Chari-6, CSV-84, Pant Chari-5 and Rajasthan Chari-1, germplasm selected from SVPUA&T, Meerut, U.P. Ten parents diallel set excluding reciprocals was made during the season of kharif 2018 by raising the crop at Crop Research Centre of Sardar Vallabhbhai Patel University of Agriculture & Technology, Meerut. All the 45 crosses and their ten parents were grown during kharif season 2019. All the 45 F1’s hybrids along with ten parents were sown in randomized block design with three replications. Observations were recorded on days to 50% flowering, plant height, leaf breadth, leaf length, stem girth, leaves per plant, leaf area, leaf stem ratio, total soluble solids and green fodder yield. Analysis of variance for parents and crosses showed highly differed significantly for all the characters. Parents vs crosses exhibited highly significant for the attributes namely, days to 50% flowering, plant height, leaf breadth, leaf length, leaf area and green fodder yield, which indicated that wide genotypic differences among the parental lines and F1’s hybrids. High heritability coupled with high genetic advance as percent of mean was observed for the traits viz., plant height, and leaves per plant, leaf area, stem girth, leaf stem ratio and green fodder yield, suggesting that the genes governing these characters may have additive effect.

On the Restricted Poisson Ridge Regression Estimator

For modeling count data, the Poisson regression model is widely used in which the response variable takes non-negative integer values. However, the presence of strong correlation between the explanatory variables causes the problem of multicollinearity. Due to multicollinearity, the variance of the maximum likelihood estimator (MLE) will be inflated causing the parameters estimation to become unstable. Multicollinearity can be tackled by using biased estimators such as the ridge estimator in order to minimize the estimated variance of the regression coefficients. An alternative approach is to specify exact linear restrictions on the parameters in addition to regression model. In this paper, the restricted Poisson ridge regression estimator (RPRRE) is suggested to handle multicollinearity in Poisson regression model with exact linear restrictions on the parameters. In addition, the conditions of superiority of the suggested estimator in comparison to some existing estimators are discussed based on the mean squared error (MSE) matrix criterion. Moreover, a simulation study and a real data application are provided to illustrate the theoretical results. The results indicate that the suggested estimator, RPRRE, outperforms the other existing estimators in terms of scalar mean squared error (SMSE). Therefore, it is recommended to use the RPRRE for the Poisson regression model when the problem of multicollinearity is present.

Genetic parameters of the type traits of Holstein-Friesian primiparous dairy cows

The aim of this study was to determine the level and variability of type traits in primiparous Holstein-Friesian cows, as well as to assess the additive genetic variance of type traits and heritability coefficients. The research was conducted on a data set of linearly evaluated primiparous Holstein-Friesian cows, which is ceded by the Main dairy cattle breed organization in the Province of Vojvodina, which is a unit of the Department of Animal Science, at the Faculty of Agriculture in Novi Sad. Data of 24226 primiparous Holstein-Friesian cows, refer to the period 2012-2015, were analysed. The phenotypic variability of the type traits, as well as effects of systemic factors affecting these traits, were analyzed using standard statistical methods in the Statistica software 13.2, and the general linear model (GLM). A Statistically highly significant effect for most of the type traits had the farm size, interaction of the year and the season of evaluation, age at linear evaluation, lactation stage and a genetic group formed by the year and country of bull/sires birth. The total type traits score (final score), which represents an evaluation of the overall primiparous cow exterior, is calculated for the 24226 linearly evaluated primiparous cows. Estimation of variance components and heritability coefficient calculation is done in the software package WOMBAT. Heritability coefficient values for the primiparous cow's type traits in the Province of Vojvodina were low to middle, ranged from 0.08 for the rear legs - side view up to 0.38 for the stature. Due to the fact that the value of the heritability coefficient depends on the variability of the traits and previous selection, lower heritability may be a consequence of the low genetic variability, or, a higher degree of variability which is a consequence of the effect of the systematic factors. Heritability coefficients indicate a genetic variability of the type traits, and its evaluation is important for the breeding value estimation for the type traits.

Two seemingly paradoxical results in linear models: the variance inflation factor and the analysis of covariance

Abstract A result from a standard linear model course is that the variance of the ordinary least squares (OLS) coefficient of a variable will never decrease when including additional covariates into the regression. The variance inflation factor (VIF) measures the increase of the variance. Another result from a standard linear model or experimental design course is that including additional covariates in a linear model of the outcome on the treatment indicator will never increase the variance of the OLS coefficient of the treatment at least asymptotically. This technique is called the analysis of covariance (ANCOVA), which is often used to improve the efficiency of treatment effect estimation. So we have two paradoxical results: adding covariates never decreases the variance in the first result but never increases the variance in the second result. In fact, these two results are derived under different assumptions. More precisely, the VIF result conditions on the treatment indicators but the ANCOVA result averages over them. Comparing the estimators with and without adjusting for additional covariates in a completely randomized experiment, I show that the former has smaller variance averaging over the treatment indicators, and the latter has smaller variance at the cost of a larger bias conditioning on the treatment indicators. Therefore, there is no real paradox.

Estimate of variance of coefficients of disturbance aerodynamic pitching and yawing moments of a rotary body with small random surface distortions of a composite thermal protection coating at supersonic flow

Estimate of variance of coefficients of disturbance aerodynamic pitching and yawing moments of a rotary body with small random surface distortions of a composite thermal protection coating at supersonic flow

Comparison of Factor Models in Capturing Factors Behind Stock Returns

I compare the five-factor model of Fama and French (2015) and the q5-factor model of Hou et al. (2018a) and find that the five-factor model outperforms the q5-factor model in terms of having closer-to-zero estimated error terms, time-series and cross-sectional covariance-to-variance ratios, and time-series variance of coefficient in most of the cases. The results thus suggest that the explanatory variables of the five-factor model are more accurate proxies of the factors stock returns depend on.

Faulty Line Identification Method Based on Bayesian Optimization for Distribution Network

Selective ground-fault protection is greatly valued for the safe and reliable operation of power systems. With the wide adoption of fault indicator in distribution network, the amount of available fault data increases dramatically. The in-depth investigation of fault recording data helps improve the accuracy of faulty line identification. To perform fault data analysis with higher efficiency, a single-phase-to-ground fault identification model based on the k-Nearest Neighbor (kNN) classification algorithm is proposed in the paper. In this model, the eigenvectors consist of wavelet energy ratio, wavelet coefficients variance and wavelet power obtained by the decomposition of transient components. Furthermore, through the theoretical analysis and experimental comparison of three parameter adjustment algorithms, Bayesian Optimization algorithm is selected to find the optimal parameters of fault identification model, and realize the adaptive adjustment of model parameters. Finally, the validity and feasibility of the model are verified by the experimental data, and the accuracy and efficiency of fault identification are improved by using Bayesian Optimization algorithm.

The LMS-Type Adaptive Filter Based on the Gaussian Model for Controlling the Variances of Coefficients

The LMS-Type Adaptive Filter Based on the Gaussian Model for Controlling the Variances of Coefficients

Phenotypic variation and estimation of genetic parameters for plant growth, fruit quality traits and bacterial blight disease resistance in gamma () irradiated seed derived progenies and germplasms of pomegranate (Punica granatum L.)

In order to improve fruit quality traits in a breeding programme, both genetic and environmental effects need to be evaluated. The partitioning of variance and genetic parameters for fruit quality traits and disease resistance were estimated in 475 accessions of pomegranate by considering the effects of genotype, environment and their interactions. The results showed that genotype contributed most to the variance except to leaf thickness, flower length, flower breadth, number of arils per fruit and seed length, Genotype and environment interactions were significant for all traits except for plant spread, leaf thickness, fruit length and rind thickness indicating the influence of environment on the attributes of pomegranate fruit. High heritability, along with genetic advances in days to maturity of fruit, Anthocyanin content, phenolic content, rind thickness, volume of juice, plant spread, plant height and acidity suggested that there were significant additive gene effects for such traits. In general, the genotypic co-efficient variances were lower in magnitude than the corresponding phenotypic co-efficient variances for all traits. The estimates of PCV and GCV were high for fruit weight, total aril weight, number of arils per fruit, volume of juice, hundred aril weight, single aril weight, acidity, anthocyanin, phenolic content and bacterial blight on fruits, indicating greater scope for improvement of these characters by simple selection. High heritability and high genetic advance as percent of mean was observed for days to maturity, plant spread, volume of juice, rind thickness, acidity, phenolic content, anthocyanin content bacterial blight on fruit, stem and disease severity.

Assessing the Robustness of Pan Evaporation Models for Estimating Reference Crop Evapotranspiration during Recalibration at Local Conditions

A classic method for assessing the reference crop evapotranspiration (ETo) is the pan evaporation (Epan) method that uses Epan measurements and pan coefficient (kp) models, which can be functions of relative humidity (RH), wind speed (u2), and temperature (T). The aim of this study is to present a methodology for evaluating the robustness of regression coefficients associated to climate parameters (RH, u2, and T) in pan method models during recalibration at local conditions. Two years of daily data from April to October (warm season) of meteorological parameters, Epan measurements from class A pan evaporimeter and ETo estimated by ASCE-standardized method for the climatic conditions of Thessaloniki (Greece, semi-arid environment), were used. The regression coefficients of six general nonlinear (NLR) regression Epan models were analyzed through recalibration using a technique called “random cross-validation nonlinear regression RCV-NLR” that produced 1000 random splits of the initial dataset into calibration and validation sets using a constant proportion (70% and 30%, respectively). The variance of the regression coefficients was analyzed based on the 95% interval of the highest posterior density distribution. NLR models that included coefficients with a 95% HPD interval that fluctuates in both positive and negative values were considered nonrobust. The machine-learning technique of random forests (RF) was also used to build a RF model that includes Epan, u2, RH, and T parameters. This model was used as a benchmark for evaluating the predictive accuracy of NLR models but, also, for assessing the relative importance of the predictor climate variables if they were all included in one NLR model. The findings of this study indicated that locally calibrated NLR functions that use only the Epan parameter presented better results, while the inclusion of additional climate parameters was redundant and led to underfitting.

Accounting for Variability in ULF Wave Radial Diffusion Models

AbstractMany modern outer radiation belt models simulate the long‐time behavior of high‐energy electrons by solving a three‐dimensional Fokker‐Planck equation for the drift‐ and bounce‐averaged electron phase space density that includes radial, pitch‐angle, and energy diffusion. Radial diffusion is an important process, often characterized by a deterministic diffusion coefficient. One widely used parameterization is based on the median of statistical ultralow frequency (ULF) wave power for a particular geomagnetic index Kp. We perform idealized numerical ensemble experiments on radial diffusion, introducing temporal and spatial variability to the diffusion coefficient through stochastic parameterization, constrained by statistical properties of its underlying observations. Our results demonstrate the sensitivity of radial diffusion over a long time period to the full distribution of the radial diffusion coefficient, highlighting that information is lost when only using median ULF wave power. When temporal variability is included, ensembles exhibit greater diffusion with more rapidly varying diffusion coefficients, larger variance of the diffusion coefficients and for distributions with heavier tails. When we introduce spatial variability, the variance in the set of all ensemble solutions increases with larger spatial scales of variability. Our results demonstrate that the variability of diffusion affects the temporal evolution of phase space density in the outer radiation belt. We discuss the need to identify important temporal and length scales to constrain variability in diffusion models. We suggest that the application of stochastic parameterization techniques in the diffusion equation may allow the inclusion of natural variability and uncertainty in modeling of wave‐particle interactions in the inner magnetosphere.

Use of Discrete Wavelet Transform to Assess Impedance Fluctuations Obtained from Cellular Micromotion.

Electric cell–substrate impedance sensing (ECIS) is an attractive method for monitoring cell behaviors in tissue culture in real time. The time series impedance fluctuations of the cell-covered electrodes measured by ECIS are the phenomena accompanying cellular micromotion as cells continually rearrange their cell–cell and cell–substrate adhesion sites. Accurate assessment of these fluctuations to extract useful information from raw data is important for both scientific and practical purposes. In this study, we apply discrete wavelet transform (DWT) to analyze the concentration-dependent effect of cytochalasin B on human umbilical vein endothelial cells (HUVECs). The sampling rate of the impedance time series is 1 Hz and each data set consists of 2048 points. Our results demonstrate that, in the Daubechies (db) wavelet family, db1 is the optimal mother wavelet function for DWT-based analysis to assess the effect of cytochalasin B on HUVEC micromotion. By calculating the energy, standard deviation, variance, and signal magnitude area of DWT detail coefficients at level 1, we are able to significantly distinguish cytotoxic concentrations of cytochalasin B as low as 0.1 μM, and in a concentration-dependent manner. Furthermore, DWT-based analysis indicates the possibility to decrease the sampling rate of the micromotion measurement from 1 Hz to 1/16 Hz without decreasing the discerning power. The statistical measures of DWT detail coefficients are effective methods for determining both the sampling rate and the number of individual samples for ECIS-based micromotion assays.

Electrodynamic Analysis of a 1D Array with Random Parameters

An electrodynamic method is proposed for analysis of infinite 1D arrays with random parameters. The approach is based on methods of compensating source and generalized scattering matrix that make it possible to analyze infinite inhomogeneous antenna arrays containing defects of periodic structure (elements with parameters that differ from the parameters of the regular structure). The array is analyzed on the assumption of small deviations of its parameters from mean values. In such an approximation, variances of amplitudes of the waves reflected from the array inputs and variance of incoherent radiation field are estimated. Quasi-periodic excitation of the array is considered. The proposed method is used for analysis of an array consisting of strip dipoles with random lengths. A dependence of the variance of reflection coefficient of the array on its parameters is obtained. A theoretical estimate of the variance coincides with the result of the Monte Carlo analysis that is used to study a finite-length dipole array.

Incremental Diversity Order for Characterization of FSO Communication Systems Over Lognormal Fading Channels

Free space optical (FSO) communication systems experience atmospheric turbulence-induced fading, which is typically modelled by lognormal distribution. Since the decreasing rate of error probability over lognormal fading channels does not converge to a finite value, the conventional definition of diversity order yields infinity and does not provide any insight into system design. In this letter, we propose the so-called incremental diversity order (IDO), which is defined as the incremental change in the decreasing rate of bit error rate (BER) on a log-log scale. As an example, we consider multiple-input multiple-output (MIMO) FSO system over lognormal turbulence channels. We derive a novel closed-form BER expression in terms of Lambert-W function and obtain IDO as a function of the number of laser sources, the number of photodetectors and the variance of the log-amplitude coefficient. We further present numerical results to validate the derived BER expression and associated IDO.

The Limits to Estimating Population-Genetic Parameters with Temporal Data.

The ability to obtain genome-wide sequences of very large numbers of individuals from natural populations raises questions about optimal sampling designs and the limits to extracting information on key population-genetic parameters from temporal-survey data. Methods are introduced for evaluating whether observed temporal fluctuations in allele frequencies are consistent with the hypothesis of random genetic drift, and expressions for the expected sampling variances for the relevant statistics are given in terms of sample sizes and numbers. Estimation methods and aspects of statistical reliability are also presented for the mean and temporal variance of selection coefficients. For nucleotide sites that pass the test of neutrality, the current effective population size can be estimated by a method of moments, and expressions for its sampling variance provide insight into the degree to which such methodology can yield meaningful results under alternative sampling schemes. Finally, some caveats are raised regarding the use of the temporal covariance of allele-frequency change to infer selection. Taken together, these results provide a statistical view of the limits to population-genetic inference in even the simplest case of a closed population.

Ecological communities from random generalized Lotka-Volterra dynamics with nonlinear feedback.

We investigate the outcome of generalized Lotka-Volterra dynamics of ecological communities with random interaction coefficients and nonlinear feedback. We show in simulations that the saturation of nonlinear feedback stabilizes the dynamics. This is confirmed in an analytical generating-functional approach to generalized Lotka-Volterra equations with piecewise linear saturating response. For such systems we are able to derive self-consistent relations governing the stable fixed-point phase and to carry out a linear stability analysis to predict the onset of unstable behavior. We investigate in detail the combined effects of the mean, variance, and covariance of the random interaction coefficients, and the saturation value of the nonlinear response. We find that stability and diversity increases with the introduction of nonlinear feedback, where decreasing the saturation value has a similar effect to decreasing the covariance. We also find cooperation to no longer have a detrimental effect on stability with nonlinear feedback, and the order parameters mean abundance and diversity to be less dependent on the symmetry of interactions with stronger saturation.

A new approach for multiclass motor imagery recognition using pattern image features generated from common spatial patterns

This paper presents a new method that can classify multiple motor imageries and can be implemented in a realistic application because of its low computation time. The method proposes the use of pattern images, generated with the common spatial pattern (CSP) technique. The paper also suggests a new algorithm to determine the best frequency bands for optimal discrimination among the diverse motor imageries to classify. The pattern images and the state images, which represent the mental state of the user in a specific segment of time, are used to compute cross-correlation coefficients. Feature vectors, including characteristics obtained with CSP, and the mean and variance of the correlation coefficients were employed to design binary classifiers with support vector machines. In addition, the work includes a real-time simulation involving a sliding window technique. The proposed method was evaluated in four datasets: IVa, IVb and V from BCI Competition III and another provided by the software BCILAB, which compared with other state-of-the-art methods. The results that overcome surpassed the methods in these competitions and other state-of-the-art methods mentioned in this paper. The method also presents short computation time, robustness between subjects and capability to classify between multiple mental states.