Empirical evaluation of the inverse Gaussian regression residuals for the assessment of influential points

Influential analysis is the main diagnostic process to obtain reliable regression results. Same is true for the generalized linear model. The present article empirically compares the performance of different residuals of the inverse Gaussian regression model to detect the influential points. The inverse Gaussian regression model residuals are further divided into two categories, that is, standardized and adjusted residuals. Cook's distance has been computed for both of the stated residuals, and then comparison of these residuals for the detection of influential point has been carried out with the help of simulation and a chemical related data set. The simulation results show that for small dispersion, the likelihood residuals are better than others and all the adjusted forms of residuals perform identically but not better than the standardized form. While for larger dispersion, all the standardized residuals perform in the same fashion, and they are better than the likelihood residuals for detection of influential points. Copyright © 2016 John Wiley & Sons, Ltd.     

A postprocessor for a posteriori error estimation of computed flow parameters

The possibility is explored of creating a postprocessor for a posteriori error estimation of computed target functionals based on the residual generated by a high-accuracy stencil and adjoint parameters as applied to a numerical solution. The applicability of this approach to the supersonic Euler equations is confirmed by computing the density at a control point.     

Asymtotics of M-estmation in Non-linear Regression

Consider the standard non-linear regression model y_i=g(x_i,θ_0) ε_j,i=1,...,n whereg(x,θ) is a continuous function on a bounded closed region X×θ_0 is the unknown parametervector in R~p,{x_1,x_2,...,x_n} is a deterministic design of experiment and {ε_1,ε_2,...,ε_n} is asequence of independent random variables.This paper establishes the existences of M-estimates andthe asymptotic uniform linearity of M-scores in a family of non-linear regression models when theerrors are independent and identically distributed.This result is then used to obtain the asymptoticdistribution of a class of M-estimators for a large class of non-linear regression models.At the sametime,we point out that Theorem 2 of Wang (1995) (J.of Multivariate Analysis,vol.54,pp.227-238,Corrigenda.vol.55,p.350) is not correct.     

The empirical distribution function and partial sum process of residuals from a stationary arch with drift process

The weak convergence of the empirical process and partial sum process of the residuals from a stationary ARCH-M model is studied. It is obtained for and consistent estimate of the ARCH-M parameters. We find that the limiting Gaussian processes are no longer distribution free and hence residuals cannot be treated as i.i.d. In fact the limiting Gaussian process for the empirical process is a standard Brownian bridge plus an additional term, while the one for partial sum process is a standard Brownian motion plus an additional term. In the special case of a standard ARCH process, that is an ARCH process with no drift, the additional term disappears. We also study a sub-sampling technique which yields the limiting Gaussian processes for the empirical process and partial sum process as a standard Brownian bridge and a standard Brownian motion respectively.     

On ecological aspects of dynamics for zero slope regression for water pollution in Chile

Zero slope regression is an important problem in chemometrics, ranging from challenges of intercept-bias and slope ‘corrections’ in spectrometry, up to analysis of administrative data on chemical pollution in water in the region of Arica and Parinacota. Such issue is really complex and it integrates problems of optimal design, symmetry of errors, stabilization of the variability of estimators, dynamical system for errors up to an administrative data challenges. In this article we introduce a realistic approach to zero slope regression problem from dynamical point of view. Linear regression is a widely used approach for data fitting under assumption of normally distributed residuals. Many times non-normal residuals are observed and also theoretically justified. Our solution to such problem uses the recently introduced inference function called score function of distribution. As a minimization criterion, the minimum information of residuals criterion is used. The score regression appears to be a direct generalization of the least-squares regression for an arbitrary known (believed) distribution of residuals. The score estimation is also distribution sensitive version of M-estimation. The capability of the method is demonstrated by water pollution data examples.     

用加权余量法分析固支圆板和环板在Mises屈服条件下的极限荷载

应用加权余量法求出了承受线性荷载的固支圆板和承受均布荷载的内边界支承环板在 Mises屈服条件下的极限荷载的近似值 ,并与最大弯矩极限条件结果进行了比较 ,说明本文计算结果较合理。     

Two notes on nonlocal field theory

In this paper, two fundamental problems completely unsolved in nonlocal field theory are studied. The first is the dependence of nonlocal residuals. By studying this problem, a theorem concerning the relationship between the residuals of nonlocal body force and nonlocal moment of momentum is given and proven. The other problem is how to give the stress boundary conditions in the linear theory of nonlocal elasticity. The stress boundary conditions obtained in this paper can not only answer why the nonlocal stress solution satisfying the boundary conditionst _ji ^(s) n _j ¦O ₂ =p _i (p _i is a constant) on the surface of crack does not exist but also give a model of the molecular cohesive stress on the crack tip.     

简支方板在弹性基础上的塑性动力响应

用加权残值法中的矩量法解简支方板塑性动力响应，分析弹性基础对方板塑性动力响应的影响，解出了弹性基础上方板残余挠度和终止运动时间。     

结构分析的样条函数方法

本文介绍了近几年来结构分析中的各种样条函数方法及其在工程中的应用,以及样条函数方法研究的新动态和笔者的最近研究成果。     

Enhanced measurement of strain distributions

A theoretical approach is presented that uses multiple strain gages to accurately measure complicated strain distributions. The technique is based on the method of weighted residuals in conjunction with measured strain data and is applicable for arbitrary in-plane strain distributions. Conventional measurements using strain gages are shown to represent a particular case of the approach presented. The experimental characterization of unidimensional strain fields is discussed in detail. Two approaches are presented; these are based on linear and quadratic approximations of the strain field. The strain distribution for two important practical problems is evaluated assuming ideal conditions to assess the performance of the proposed approach. In both cases, the simulated results demonstrate that measurement error resulting from the finite size of a strain gage may be reduced. That is, a larger strain gage may be used for a given maximum admissible error. The method also allows a minimal error of measured nonlinear strains.