TITLE

Criteria for Unconstrained Global Optimization in Nonconvex Problems

AUTHOR(S)
Demidenko, Eugene
PUB. DATE
September 2007
SOURCE
AIP Conference Proceedings;9/6/2007, Vol. 936 Issue 1, p147
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
We develop criteria for the existence and uniqueness of global minima of a continuous bounded function on a noncompact set. Special attention is given to the problem of parameter estimation via minimization of the sum of squares in nonlinear regression and maximum likelihood. Definitions of local convexity and unimodality are given using the level set. A fundamental theorem of nonconvex optimization is formulated: If a function approaches the minimal limiting value at the boundary of the optimization domain from below and its Hessian matrix is positive definite at the point where the gradient vanishes then the function has a unique minimum. It is shown that the local convexity level of the sum of squares is equal to the minimal squared radius of the regression curvature.
ACCESSION #
26501835

 

Related Articles

  • Unbiased Quasi-regression*. Guijun Yang; Lu Lin; Runchu Zhang // Chinese Annals of Mathematics;Apr2007, Vol. 28 Issue 2, p177 

    Quasi-regression, motivated by the problems arising in the computer experiments, focuses mainly on speeding up evaluation. However, its theoretical properties are unexplored systemically. This paper shows that quasi-regression is unbiased, strong convergent and asymptotic normal for parameter...

  • Robust Estimation Algorithm for Technological Parameters of Automotive Ignition Coil. Qisong Wang; Dan Liu; Yongping Zhao // International Journal of Digital Content Technology & its Applic;Dec2011, Vol. 5 Issue 12, p119 

    An Improved Robust Least Squares Support Vector Regression (IRLSSVR) based on IGGIII weight function is presented to solve the problem that the technological parameters estimated precision of automotive ignition coil are usually influenced by gross errors. By utilizing IGGIII weight function,...

  • Neural Error Regression Diagnosis (NERD): A Tool for Model Bias Identification and Prognostic Data Assimilation. Abramowitz, Gab; Gupta, Hoshin; Pitman, Andy; Wang, Yingping; Leuning, Ray; Cleugh, Helen; Hsu, Kuo-lin // Journal of Hydrometeorology;Feb2006, Vol. 7 Issue 1, p160 

    Data assimilation in the field of predictive land surface modeling is generally limited to using observational data to estimate optimal model states or restrict model parameter ranges. To date, very little work has attempted to systematically define and quantify error resulting from a model's...

  • A paradox concerning nuisance parameters and projected estimating functions. Henmi, Masayuki; Eguchi, Shinto // Biometrika;Dec2004, Vol. 91 Issue 4, p929 

    This paper is concerned with a paradox associated with parameter estimation in the presence of nuisance parameters. In a statistical model with unknown nuisance parameters, the efficiency of an estimator of a parameter usually increases when the nuisance parameters are known. However the...

  • Alternative Formulations for Incorporating Lateral Boundary Data into Limited-Area Models. Tudor, Martina; Termonia, Piet // Monthly Weather Review;Jul2010, Vol. 138 Issue 7, p2867 

    Limited-area models (LAMs) use higher resolutions and more advanced parameterizations of physical processes than global numerical weather prediction models, but suffer from one additional source of error—the lateral boundary condition (LBC). The large-scale model passes the information on...

  • A NOTE ON TWO CURIOSITIES IN LINEAR REGRESSION. Poirier, Dale J. // American Economist;Spring78, Vol. 22 Issue 1, p77 

    Focuses on two relationships involving alternative estimators of beta in linear regression. Condition for the operationality of the linear estimator of beta with minimum mean square error; Approaches in dealing with unknown parameters.

  • Normal Forms in One-Parametric Optimization. Davydov, A. A.; Jongen, H. Th. // Annals of Operations Research;2001, Vol. 101 Issue 1-4, p255 

    We deal with one-parametric families of optimization problems in finite dimension with a finite number of (in-)equality constraints. Its set of generalized critical points can be classified according to five types (cf. [3]). In this paper the corresponding normal forms of the problem near each...

  • Constraining Land Surface and Atmospheric Parameters of a Locally Coupled Model Using Observational Data. Liu, Yuqiong; Gupta, Hoshin V.; Sorooshian, Soroosh; Bastidas, Luis A.; Shuttleworth, William J. // Journal of Hydrometeorology;Apr2005, Vol. 6 Issue 2, p156 

    In coupled land surface�atmosphere modeling, the possibility and benefits of constraining model parameters using observational data bear investigation. Using the locally coupled NCAR Single-column Community Climate Model (NCAR SCCM), this study demonstrates some feasible, effective...

  • MAXIMUM LIKELIHOOD ESTIMATION OF CO-CHANNEL MULTICOMPONENT POLYNOMIAL PHASE SIGNALS USING IMPORTANCE SAMPLING. Cheng, H.; Zeng, D.; Zhu, J.; Tang, B. // Progress in Electromagnetics Research C;Nov2011, Vol. 23, p111 

    Unlike some traditional polynomial phase signal (PPS) parameter estimation methods restricted to monocomponent case, this paper focuses on the parameter estimation of multicomponent PPSs mixed in a single channel, which is more sophisticated and always involves the cross-term issue. In this...

Share

Read the Article

Courtesy of VIRGINIA BEACH PUBLIC LIBRARY AND SYSTEM

Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics