Regression with Stable Errors Based on Order Statistics

Classical regression approaches are not robust when errors are heavy-tailed or asymmetric. That may be due to the non-existence of the mean or variance of the error distribution. Estimation based on trimmed data, which ignored outlier or leverage points, has an old history and frequently used. This procedure chooses fixed cut-off points. In this work, we use this idea recently applied for initial estimates of regression coefficients with heavy-tailed stable errors. We propose an effective procedure to calculate the cut-off points based on the tail index and skewness parameters of errors. We use the property of the existence of some moments of stable distribution order statistics. Data are trimmed based on ordered residuals of a least square regression. However, the trimmed data’s optimal number is determined based on the number of error order statistics whose variance exists. Then, we use the rest of the ordered data to estimate the regression coefficients. Based on these order statistics’ joint distribution, we analytically compute the bias and variance of the introduced estimator of regression parameters that was impossible for regression with stable errors.

Numerical Analysis of Double Integral of Trigonometric Function Using Romberg Method

In general, solving the two-fold integral of trigonometric functions is not easy to do analytically. Therefore, we need a numerical method to get the solution. Numerical methods can only provide solutions that approach true value. Thus, a numerical solution is also called a close solution. However, we can determine the difference between the two (errors) as small as possible. Numerical settlement is done by consecutive estimates (iteration method). The numerical method used in this study is the Romberg method. Romberg's integration method is based on Richardson's extrapolation expansion, so that there is a calculation of the integration of functions in two estimating ways I (h1) and I (h2) resulting in an error order on the result of the completion increasing by two, so it needs to be reviewed briefly about how the accuracy of the method. The results of this study indicate that the level of accuracy of the Romberg method to the analytical method (exact) will give the same value, after being used in several simulations.

Analysis of the Morley element for the Cahn–Hilliard equation and the Hele-Shaw flow

The paper analyzes the Morley element method for the Cahn–Hilliard equation. The objective is to prove the numerical interfaces of the Morley element method approximate the Hele-Shaw flow. It is achieved by establishing the optimal error estimates which depend on 1/ε polynomially, and the error estimates should be established from lower norms to higher norms progressively. If the higher norm error bound is derived by choosing test function directly, we cannot obtain the optimal error order, and we cannot establish the error bound which depends on 1/ε polynomially either. Different from the discontinuous Galerkin (DG) space [Feng et al. SIAM J. Numer. Anal. 54 (2016) 825–847], the Morley element space does not contain the finite element space as a subspace such that the projection theory does not work. The enriching theory is used in this paper to overcome this difficulty, and some nonstandard techniques are combined in the process such as the a priori estimates of the exact solution u, integration by parts in space, summation by parts in time, and special properties of the Morley elements. If one of these techniques is lacked, either we can only obtain the sub-optimal piecewise L∞(H2) error order, or we can merely obtain the error bounds which are exponentially dependent on 1/ε. Numerical results are presented to validate the optimal L∞(H2) error order and the asymptotic behavior of the solutions of the Cahn–Hilliard equation.

Modeling of the elastic characteristics of a long-fiber reinforced composite with an arbitrary orientation of the reinforcing fibers

Currently, the question of the best scheme for constructing a mathematical model of a homogeneous anisotropic elastic material equivalent to fiber reinforced plastic composite (FRP) with arbitrary laminate stacking sequence configuration, have been remained open. A new method for the theoretical prediction of anisotropic elastic characteristics of material equivalent to a given FRP is suggested in this paper. Results are obtained for representative volume elemen (RV) which has been cut out in three different ways from FRP. Calculations for a specific FRP have shown that the FRP replacement by a homogeneous anisotropic material equivalent to it leads to an error order 10% for the elastic properties.

Stochastic Maximum Principle of Near-Optimal Control of Fully Coupled Forward-Backward Stochastic Differential Equation

This paper first makes an attempt to investigate the near-optimal control of systems governed by fully nonlinear coupled forward-backward stochastic differential equations (FBSDEs) under the assumption of a convex control domain. By Ekeland’s variational principle and some basic estimates for state processes and adjoint processes, we establish the necessary conditions for anyε-near optimal control in a local form with an error order of exactε1/2. Moreover, under additional convexity conditions on Hamiltonian function, we prove that anε-maximum condition in terms of the Hamiltonian in the integral form is sufficient for near-optimality of orderε1/2.

Further Comparisons of Finite Difference Schemes for Computational Modelling of Biosensors

Simulations are presented for a reaction-diffusion system within a thin layer containing an enzyme, fed with a substrate from the surrounding electrolyte. The chemical term is of the nonlinear Michaelis-Menten type and requires a technique such as Newton iteration for solution. It is shown that approximating the nonlinear chemical term in these systems by a linearised form reduces both the accuracy and, in the case of second-order methods such as Crank-Nicolson, reduces the global error order from O(δT2) to O(δT). The first-order methods plain backwards implicit with and without linearisation, and Crank-Nicolson with linearisation are all of O(δT) and very similar in performance, requiring, for a given accuracy target, an order of magnitude more CPU time than the efficient methods backward implicit with extrapolation and Crank-Nicolson, both with Newton iteration to handle the nonlinearity. Steady state computations agree with expectations, tending to the known solutions for limiting cases. The Crank-Nicolson method shows some concentration oscillations close to the outer layer boundary but this does not propagate to the inner boundary at the electrode. The backward implicit methods do not result in such oscillations and if concentration profiles are of interest, may be preferred.

Poisson Approximation in a Poisson Limit Theorem Inspired by Coupon Collecting

In this paper we refine a Poisson limit theorem of Gnedenko and Kolmogorov (1954): we determine the error order of a Poisson approximation for sums of asymptotically negligible integer-valued random variables that converge in distribution to the Poisson law. As an application of our results, we investigate the case of the coupon collector's problem when the distribution of the collector's waiting time is asymptotically Poisson.

Poisson Approximation in a Poisson Limit Theorem Inspired by Coupon Collecting

In this paper we refine a Poisson limit theorem of Gnedenko and Kolmogorov (1954): we determine the error order of a Poisson approximation for sums of asymptotically negligible integer-valued random variables that converge in distribution to the Poisson law. As an application of our results, we investigate the case of the coupon collector's problem when the distribution of the collector's waiting time is asymptotically Poisson.

Radiative transfer: exact Rayleigh scattering series and a formula for daylight

Daylight, or sky light, is sunlight Rayleigh scattered by the atmosphere onto the ground. This random scattering propagation through clear air is governed by ‘radiative transfer’. Beyond the single-scattering approximation, the famous virtuoso analysis of Chandrasekhar formulated the problem and offered exact, but rather involved, and ultimately numerical, algorithms for its solution. However, there is no real difficulty in writing down directly the exact Rayleigh scattering series in integrals. Its practical utility is limited to fairly small thicknesses T of atmosphere (compared with the mean free path), but the Earth has just such. Here even the next order beyond single scattering (error order T 2 ) supplies a formula for the brightness and partial polarization of daylight across the sky, which captures the essential topology of the polarization pattern, and also remains uniformly valid in the small thickness limit, for all elevations of the Sun and viewing angles. The status of the mathematical polarization direction pattern invented by Berry, Dennis and Lee as the simplest fit to the required topology is clarified.