- Published 2015

In this talk, we consider white noise testing and model diagnostic checking for stationary functional time series. To test for the functional white noise null hypothesis, we propose a Cramer-von Mises type test based on the functional periodogram introduced by Panaretos and Tavakolithe (2013). Using the Hilbert space approach, we derive the asymptotic distribution of the test statistic under suitable assumptions. A new block bootstrap procedure is introduced to obtain the critical values from the non-pivotal limiting distribution. Compared to existing methods, our approach is robust to the dependence within white noise and it does not involve the choices of functional principal components and lag truncation number. We employ the proposed method to check the adequacy of functional linear models and functional autoregressive models of order one by testing the uncorrelatedness of the residuals. Monte Carlo simulations are provided to demonstrate the empirical advantages of the proposed Spring 2015 UMKC Math & Stats Colloquium Series 2 method over existing alternatives. Our method is illustrated via an application to cumulative intradaily returns. Friday, March 20, 2015 Dr. Paul Rulis Assistant Professor, Department of Physics The University of Missouri—Kansas City Can Artificial Neural Networks Be Used to Supplant Self-Consistent Field Calculations within Density Functional Theory? ABSTRACT Ab initio electronic structure calculations based on quantum mechanics have become essential tools for materials scientists that need to access wave function-based material properties. Although advanced methods and advanced computers have increased the size of the material systems that can be studied with these methods, it has proved difficult to scale beyond a few thousand atoms. This is a point of frustration because many of the more interesting material systems at the nano-scale require on the order of ten to twenty thousand atoms to model. Similarly, if a problem requires the use of ab initio molecular dynamics it will be severely restricted in its duration because of the computational cost. In this presentation we explore an alternative method for calculating accurate total energies of complex defect containing solids that is based on machine learning instead of traditional self-consistent field (SCF) calculations. Progress of the method as applied to a passive defect model in silicon, a self-interstitial model in silicon, and a model of amorphous silica will be presented.Ab initio electronic structure calculations based on quantum mechanics have become essential tools for materials scientists that need to access wave function-based material properties. Although advanced methods and advanced computers have increased the size of the material systems that can be studied with these methods, it has proved difficult to scale beyond a few thousand atoms. This is a point of frustration because many of the more interesting material systems at the nano-scale require on the order of ten to twenty thousand atoms to model. Similarly, if a problem requires the use of ab initio molecular dynamics it will be severely restricted in its duration because of the computational cost. In this presentation we explore an alternative method for calculating accurate total energies of complex defect containing solids that is based on machine learning instead of traditional self-consistent field (SCF) calculations. Progress of the method as applied to a passive defect model in silicon, a self-interstitial model in silicon, and a model of amorphous silica will be presented.

@inproceedings{Wang2015Spring2U,
title={Spring 2015 UMKC Math & Stats Colloquium},
author={Xing-he Wang},
year={2015}
}