Recursive algorithms are very useful for computing M-estimators of regression coefficients and scatter parameters. In this article, it is shown that for a nondecreasing ul （t）, under some mild conditions the recursi...Recursive algorithms are very useful for computing M-estimators of regression coefficients and scatter parameters. In this article, it is shown that for a nondecreasing ul （t）, under some mild conditions the recursive M-estimators of regression coefficients and scatter parameters are strongly consistent and the recursive M-estimator of the regression coefficients is also asymptotically normal distributed. Furthermore, optimal recursive M-estimators, asymptotic efficiencies of recursive M-estimators and asymptotic relative efficiencies between recursive M-estimators of regression coefficients are studied.展开更多
Simple linear regression analysis has been used to map QTL for quantitative traits. Many traits of biological interest and/or economical importance in various species show binary phenotypic distributions (e.g., presen...Simple linear regression analysis has been used to map QTL for quantitative traits. Many traits of biological interest and/or economical importance in various species show binary phenotypic distributions (e.g., presence or absence). It has been shown that such a binary trait also can be analyzed with the simple linear regression, subject to virtually no loss in power compared to the generalized linear model analysis. Binary trait is a special case of a multiple categorical trait (e.g., low, medium or high). We propose a mechanism to decompose a multiple categorical trait into an array of correlated binary variables. The categorical trait turned multiple binary traits are analyzed with a multivariate linear regression method. Turning the problem of categorical trait mapping into that of multivariate mapping allows the exploration of pleiotropic effects of QTL for different categories. Efficiency of the method is verified through a series of simulation experiments.展开更多
Productivity is a very important element in the process of construction project management especially with regard to the estimation of the duration of the construction activities, this study aims at developing constru...Productivity is a very important element in the process of construction project management especially with regard to the estimation of the duration of the construction activities, this study aims at developing construction productivity estimating model for marble finishing works of floors using Multivariable Linear Regression technique (MLR). The model was developed based on 100 set of data collected in Iraq for different types of projects such as residential, commercial and educational projects. Which these are used in developing the model and evaluating its performance. Ten influencing factors are utilized for productivity forecasting by MLR model, and they include age, experience, number of the assist labor, height of the floor, size of the marbles tiles, security conditions, health status for the work team, weather conditions, site condition, and availability of construction materials. One model was built for the prediction of the productivity of marble finishing works for floors. It was found that MLR have the ability to predict the productivity for finishing works with excellent degree of accuracy of the coefficient of correlation (R) 90.6%, and average accuracy percentage of 96.3%. This indicates that the relationship between the independent and independent variables of the developed models is good and the predicted values from a forecast model fit with the real-life data.展开更多
In current paper, a quantitative structure-activity relationship （QSAR） study was performed for the prediction of acute toxicity of aromatic amines. A set of 56 compounds was randomly divided into a training set of ...In current paper, a quantitative structure-activity relationship （QSAR） study was performed for the prediction of acute toxicity of aromatic amines. A set of 56 compounds was randomly divided into a training set of 46 compounds and a test set of 10 compounds. The electronic and topological descriptors computed by the Scigress package and Dragon software were used as predictor variables. Multiple linear regression （MLR） and support vector machine （SVM） were utilized to build the linear and nonlinear QSAR models, respectively. The obtained models with five descriptors show strong predictive ability. The linear model fits the training set with R2 = 0.71, with higher SVM values of R2 = 0.77. The validation results obtained from the test set indicate that the SVM model is comparable or superior to that obtained by MLR, both in terms of prediction ability and robustness.展开更多
Efficiency and precision in prediction of Chlorophyll-a using this model is still a pandemic among researchers, due to the natural conditions in ocean water systems itself, which involved chemical, biological and phys...Efficiency and precision in prediction of Chlorophyll-a using this model is still a pandemic among researchers, due to the natural conditions in ocean water systems itself, which involved chemical, biological and physical processes and interaction among them may affect the model performance drastically. Thus, to overcome this problem as well as to improve the strength of MLR, we proposed a hybrid approach, i.e., an Artificial Neural Network to the MLR coins as Artificial Neural Network-Multiple Linear Regression (ANN-MLR). To investigate the performance of the proposed model, we compared Multiple Linear Regression (MLR), Artificial Neural Network (ANN) and proposed hybrid Artificial Neural Network and Multiple Linear Regression (ANN-MLR) in the prediction of chlorophyll-a (chl-a) concentration by statistical measurement which are MSE and MAE. Achieving our objectives of study, we used 4 parameters, i.e. temperature (°C), pH, salinity (ppt), DO (ppm) at the Offshore Kuala Terengganu, Terengganu, Malaysia. The results showed that our proposed model can improve the performance of the model as compared to ANN and MLR due to small errors generated, error reduced, and increased the correlation coefficient for all parameters in both MSE and MAE, respectively. Thus, this result indicated that our proposed model is efficient, precise and almost perfect correlation as compared to ANN and MLR.展开更多
目的:多元线性回归模型在保持输入自变量光谱信息和空间特征的同时,通过线性变换获取自变量和因变量的光谱拟合关系,对原输入自变量的光谱信息进行优化,从而获得高空间分辨率和丰富光谱信息的重构数据。方法:利用同期获取的OLI(Operatio...目的:多元线性回归模型在保持输入自变量光谱信息和空间特征的同时,通过线性变换获取自变量和因变量的光谱拟合关系,对原输入自变量的光谱信息进行优化,从而获得高空间分辨率和丰富光谱信息的重构数据。方法:利用同期获取的OLI(Operational Land Imager)和PMS(Panchromatic and Multispectral Scanner)多光谱遥感影像,根据最小二乘法构建多元线性回归模型,重构生成具有丰富光谱特征和空间特征的遥感影像,从主客观两个方面评价重构影像的质量。结果:在目视解译(主观)方面,重构影像在一定程度上保留了原OLI影像的光谱特性,提升了原PMS影像的清晰度和分辨性;在量化角度(客观)方面,重构影像的信息量和平均梯度比原OLI对应波段影像的信息量(在部分波段上)和平均梯度要低,但比原PMS影像的信息量和平均梯度要高,可见重构影像的质量介于原PMS影像和OLI影像的质量之间。结论:以青海省门源回族自治县的耕地内不同作物为实例对象,利用最大似然法获取门源县青稞和油菜的空间分布,研究区实测数据验证表明,重构影像对耕地内部青稞与油菜的提取精度高于原PMS和OLI多光谱影像的提取精度。展开更多
基金supported by the Natural Sciences and Engineering Research Council of Canadathe National Natural Science Foundation of China+2 种基金the Doctorial Fund of Education Ministry of Chinasupported by the Natural Sciences and Engineering Research Council of Canadasupported by the National Natural Science Foundation of China
文摘Recursive algorithms are very useful for computing M-estimators of regression coefficients and scatter parameters. In this article, it is shown that for a nondecreasing ul （t）, under some mild conditions the recursive M-estimators of regression coefficients and scatter parameters are strongly consistent and the recursive M-estimator of the regression coefficients is also asymptotically normal distributed. Furthermore, optimal recursive M-estimators, asymptotic efficiencies of recursive M-estimators and asymptotic relative efficiencies between recursive M-estimators of regression coefficients are studied.
基金Item supported by national natural sciencefoundation( No.30471236)
文摘Simple linear regression analysis has been used to map QTL for quantitative traits. Many traits of biological interest and/or economical importance in various species show binary phenotypic distributions (e.g., presence or absence). It has been shown that such a binary trait also can be analyzed with the simple linear regression, subject to virtually no loss in power compared to the generalized linear model analysis. Binary trait is a special case of a multiple categorical trait (e.g., low, medium or high). We propose a mechanism to decompose a multiple categorical trait into an array of correlated binary variables. The categorical trait turned multiple binary traits are analyzed with a multivariate linear regression method. Turning the problem of categorical trait mapping into that of multivariate mapping allows the exploration of pleiotropic effects of QTL for different categories. Efficiency of the method is verified through a series of simulation experiments.
文摘Productivity is a very important element in the process of construction project management especially with regard to the estimation of the duration of the construction activities, this study aims at developing construction productivity estimating model for marble finishing works of floors using Multivariable Linear Regression technique (MLR). The model was developed based on 100 set of data collected in Iraq for different types of projects such as residential, commercial and educational projects. Which these are used in developing the model and evaluating its performance. Ten influencing factors are utilized for productivity forecasting by MLR model, and they include age, experience, number of the assist labor, height of the floor, size of the marbles tiles, security conditions, health status for the work team, weather conditions, site condition, and availability of construction materials. One model was built for the prediction of the productivity of marble finishing works for floors. It was found that MLR have the ability to predict the productivity for finishing works with excellent degree of accuracy of the coefficient of correlation (R) 90.6%, and average accuracy percentage of 96.3%. This indicates that the relationship between the independent and independent variables of the developed models is good and the predicted values from a forecast model fit with the real-life data.
基金Supported by the Ministry of Environmental Protection of China(No.2011467037)
文摘In current paper, a quantitative structure-activity relationship （QSAR） study was performed for the prediction of acute toxicity of aromatic amines. A set of 56 compounds was randomly divided into a training set of 46 compounds and a test set of 10 compounds. The electronic and topological descriptors computed by the Scigress package and Dragon software were used as predictor variables. Multiple linear regression （MLR） and support vector machine （SVM） were utilized to build the linear and nonlinear QSAR models, respectively. The obtained models with five descriptors show strong predictive ability. The linear model fits the training set with R2 = 0.71, with higher SVM values of R2 = 0.77. The validation results obtained from the test set indicate that the SVM model is comparable or superior to that obtained by MLR, both in terms of prediction ability and robustness.
文摘Efficiency and precision in prediction of Chlorophyll-a using this model is still a pandemic among researchers, due to the natural conditions in ocean water systems itself, which involved chemical, biological and physical processes and interaction among them may affect the model performance drastically. Thus, to overcome this problem as well as to improve the strength of MLR, we proposed a hybrid approach, i.e., an Artificial Neural Network to the MLR coins as Artificial Neural Network-Multiple Linear Regression (ANN-MLR). To investigate the performance of the proposed model, we compared Multiple Linear Regression (MLR), Artificial Neural Network (ANN) and proposed hybrid Artificial Neural Network and Multiple Linear Regression (ANN-MLR) in the prediction of chlorophyll-a (chl-a) concentration by statistical measurement which are MSE and MAE. Achieving our objectives of study, we used 4 parameters, i.e. temperature (°C), pH, salinity (ppt), DO (ppm) at the Offshore Kuala Terengganu, Terengganu, Malaysia. The results showed that our proposed model can improve the performance of the model as compared to ANN and MLR due to small errors generated, error reduced, and increased the correlation coefficient for all parameters in both MSE and MAE, respectively. Thus, this result indicated that our proposed model is efficient, precise and almost perfect correlation as compared to ANN and MLR.
文摘目的:多元线性回归模型在保持输入自变量光谱信息和空间特征的同时,通过线性变换获取自变量和因变量的光谱拟合关系,对原输入自变量的光谱信息进行优化,从而获得高空间分辨率和丰富光谱信息的重构数据。方法:利用同期获取的OLI(Operational Land Imager)和PMS(Panchromatic and Multispectral Scanner)多光谱遥感影像,根据最小二乘法构建多元线性回归模型,重构生成具有丰富光谱特征和空间特征的遥感影像,从主客观两个方面评价重构影像的质量。结果:在目视解译(主观)方面,重构影像在一定程度上保留了原OLI影像的光谱特性,提升了原PMS影像的清晰度和分辨性;在量化角度(客观)方面,重构影像的信息量和平均梯度比原OLI对应波段影像的信息量(在部分波段上)和平均梯度要低,但比原PMS影像的信息量和平均梯度要高,可见重构影像的质量介于原PMS影像和OLI影像的质量之间。结论:以青海省门源回族自治县的耕地内不同作物为实例对象,利用最大似然法获取门源县青稞和油菜的空间分布,研究区实测数据验证表明,重构影像对耕地内部青稞与油菜的提取精度高于原PMS和OLI多光谱影像的提取精度。
