检索结果-内蒙古大学图书馆

Online learning and forecast combination in unbalanced panels

ECONOMETRIC REVIEWS 2017年第1-3期36卷 257-288页

作者： Lahiri, Kajal Peng, Huaming Zhao, Yongchen SUNY Albany Dept Econ 1400 Washington Ave Albany NY 12222 USA Towson Univ Dept Econ Towson MD USA

This article evaluates the performance of a few newly proposed online forecast combination algorithms and compares them with some of the existing ones including the simple average and that of Bates and Granger (1969). We derive asymptotic results for the new algorithms that justify certain established approaches to forecast combination including trimming, clustering, weighting, and shrinkage. We also show that when implemented on unbalanced panels, different combination algorithms implicitly impute missing data differently, so that the performance of the resulting combined forecasts are not comparable. After explicitly imputing the missing observations in the U.S. Survey of Professional Forecasters (SPF) over 1968 IV-2013 I, we find that the equally weighted average continues to be hard to beat, but the new algorithms can potentially deliver superior performance at shorter horizons, especially during periods of volatility clustering and structural breaks.

关键词： Machine learning recursive algorithms SPF forecasts missing data C22 C53 C14

来源：评论

学校读者我要写书评

暂无评论

Activation-based recursive self-organising maps: A general formulation and empirical results

引用

NEURAL PROCESSING LETTERS 2006年第2期24卷 119-136页

作者： Hynna, Kevin I. Kaipainen, Mauri Aalto Univ Lab Comp & Informat Sci Neural Networks Res Ctr FI-02015 Helsinki Finland Tallinn Univ Dept Informat New Media EE-10120 Tallinn Estonia

We generalize a class of neural network models that extend the Kohonen Self-Organising Map (SOM) algorithm into the sequential and temporal domain using recurrent connections. Behaviour of the class of Activation-based recursive Self-Organising Maps (ARSOM) is discussed with respect to the choice of transfer function and parameter settings. By comparing performances to existing benchmarks we demonstrate the robustness and systematicity of the ARSOM models, thus opening the door to practical applications.

关键词： recurrent neural networks recursive algorithms representing context self-organizing maps sequential order

来源：评论

学校读者我要写书评

暂无评论

Angular and hyperangular momentum recoupling, harmonic superposition and Racah polynomials: a recursive algorithm

引用

CHEMICAL PHYSICS LETTERS 2001年第5-6期344卷 587-600页

作者： Aquilanti, V Cavalli, S Coletti, C Univ Perugia Dipartimento Chim Chim Lab I-06123 Perugia Italy

Generalized 6j symbols are defined in terms of orthonormalized Racah polynomials of a discrete variable and given explicitly as hypergeometric F-4(3)(1) series. They extend the recoupling coefficients of ordinary angular moment-am algebra, including multiples of 1/4 as quantum numbers. A three-term recurrence relationship is exploited for extensive calculations and illustration of their properties. Their role is outlined as matrix elements for superpositions (or overlaps) both between alternative spherical and hyperspherical harmonics and between alternative Sturmian sets, an important case being that of four-dimensional harmonies of S-3, which apply to the momentum-space hydrogen atom orbitals. (C) 2001 Elsevier Science B.V. All rights reserved.

关键词： Quantum number ILLUSTRATIONS hydrogen atoms recursive algorithms Angle multiples Racah coefficient recoupling discrete transforms Polynomial

来源：评论

学校读者我要写书评

暂无评论

recursive Closed-loop PARSIM-E Subspace Identification

引用

IFAC-PapersOnLine 2015年第28期48卷 880-885页

作者： Hou, Jie Chen, Fengwei Liu, Tao School of Control Science and Engineering Dalian University of Technology Dalian China

In this paper a recursive closed-loop PARSIMonious subspace identification method using modified innovation Estimation (PARSIM-E) is proposed, which can significantly improve computational efficiency for practical application. By recursively updating the product of the extended observability and controllability matrices, the non-causal parameters included in the triangular Toeplitz matrices are left out to avoid redundant estimation. A propagator is used to extract the system state in order to estimate the system matrices. The convergence of the proposed method is analyzed with a strict proof. Two illustrative examples are given to demonstrate the effectiveness of proposed algorithm. © 2015

关键词： Computational efficiency Matrix algebra State space methods Closed loop identification convergence propagator recursive algorithms State space models

来源：评论

学校读者我要写书评

暂无评论

An Algorithm to Find Common Zeroing-Output Systems of Arbitrarily Switched Linear Time-Invariant Systems

引用

IFAC-PapersOnLine 2017年第1期50卷 14867-14872页

作者： Lee, Ti-Chung Tan, Ying Mareels, Iven Department of Electrical Engineering Minghsin University of Science and Technology Hsinchu Taiwan Department of Electrical and Electronic Engineering the University of Melbourne VIC3010 Australia

This paper proposes an algorithm to find a common zeroing-output system (CZOS) that plays a key role in studying uniform global exponential stability for a class of switched linear time-invariant systems under arbitrary switching. When a quadratic common Lyapunov function is hard to find while a quadratic common weak Lyapunov function is available, the concepts of CZOS can be used to assist in checking uniform global attractivity. When a CZOS for a given switched linear time-invariant system exists, a recursive algorithm is proposed to systematically find this CZOS. One example demonstrates the effectiveness of the proposed algorithm. © 2017

关键词： System stability Invariance Linear systems Lyapunov functions Nonlinear control systems Time varying control systems Arbitrary switching Common Lyapunov functions Common weak Lyapunov function Global attractivity Linear time invariant systems recursive algorithms Uniform global exponential stabilities

来源：评论

学校读者我要写书评

暂无评论

recursive Subspace-based Predictive Control and Its Application to Fault-tolerant Control

引用

IFAC-PapersOnLine 2018年第24期51卷 696-702页

作者： Gao, Jingjing Yang, Xu Li, Linlin Luo, Hao Ding, Steven X. Zhang, Kai Peng, Kaixiang School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing100083 China School of Astronautics Harbin Institute of Technology Harbin150001 China Institute of Automatic Control and Complex Systems Faculty of Engineering University of Duisburg-Essen Duisburg47057 Germany

This paper investigates the data-driven realization of predictive controllers, and its application to fault-tolerant control. With the aid of subspace identification method, the subspace predictor is first derived in the closed-loop setup. To cope with the change in the process, a recursive algorithm is adopted to recursive update the subspace matrices. The data-driven predictive controller is designed by seeking the solution of the quadratic programming problem, without the information of the predesigned controller. Then, it is applied to fault-tolerant control by implementing the recursive subspace predictor. In the end, the effectiveness of the proposed method is verified by a case study on the three-tank system. © 2018

关键词： Controllers Fault tolerance Quadratic programming Fault tolerant control Predictive control Predictive controller Quadratic programming problems recursive algorithms Subspace identification methods Subspace predictors Three tank system

来源：评论

学校读者我要写书评

暂无评论

THE BLOCK REGULARIZED PARAMETER ESTIMATOR AND ITS PARALLELIZATION

引用

AUTOMATICA 1995年第8期31卷 1125-1136页

作者： KADLEC, J GASTON, FMF IRWIN, GW Control Engineering Research Centre Department of Electrical and Electronic Engineering The Queen's University of Belfast Belfast BT9 5AH Northern Ireland Stranmillis Road United Kingdom

Kulhavy's regularised parameter identification concept protects the adaptive recursive estimation of a linear regression model from numerical difficulties associated with standard exponential weighting in cases where the processed data is not sufficiently exciting. Unfortunately, such robustness incurs a severe penalty in computational complexity, which militates against practical applications. This paper presents a new block regularised parameter estimator that is compatible with the requirements for implementation on a parallel architecture. Owing to the accumulated regularisation in blocks, the achieved throughput of the estimator is an order of magnitude higher in comparison with the general framework of Kulhavy and more comparable to recursive least squares on a systolic array. The processing cells operate at almost 100% efficiency, and are only connected to their nearest neighbours by one-directional connections. This new parameter estimator offers significant potential for identification, adaptive filtering and adaptive control applications, particularly in the real-time domain.

关键词： recursive algorithms LEAST-SQUARES IDENTIFICATION PARALLEL PROCESSORS ROBUST ESTIMATION

来源：评论

学校读者我要写书评

暂无评论

Optimal buying at the global minimum in a regime switching model

引用

MATHEMATICAL SOCIAL SCIENCES 2016年 84卷 50-55页

作者： Yang, Aijun Liu, Yue Xiang, Ju Yang, Hongqiang Nanjing Forestry Univ Coll Econ & Management Nanjing Jiangsu Peoples R China Jiangsu Univ Sch Finance & Econ Zhenjiang Jiangsu Peoples R China South Univ Sci & Technol China Dept Finance Shenzhen Guangdong Peoples R China

This paper addresses the problem of buying an asset at its expected globally minimal price, to that end, we model it as an optimal stopping problem with regime switching driven by a continuous-time Markov chain. We characterize the optimal stopping time by optimizing the value functions and writing them as solutions of a system of integral equations. Finally we develop a stochastic recursive algorithm for numerical implementation. (C) 2016 Elsevier B.V. All rights reserved.

关键词： switching recursive algorithms Global Minimum Purchasing Stopping time system parameter optical coherence tomography angiography device Optimal

来源：评论

学校读者我要写书评

暂无评论

STOCHASTIC FRAMEWORK FOR recursive COMPUTATION OF SPLINE FUNCTIONS .2. SMOOTHING SPLINES

引用

JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS 1980年第2期30卷 255-268页

作者： WEINERT, HL BYRD, RH SIDHU, GS JOHNS HOPKINS UNIV DEPT MATH SCIBALTIMOREMD 21218 UNIV NACL AUTONOMA MEXICO INST INVEST MATEMAT APLICADES & SISTEMASMEXICO CITY 20MEXICO

Many of the optimal curve-fitting problems arising in approximation theory have the same structure as certain estimation problems involving random processes. We develop this structural correspondence for the problem of smoothing inaccurate data with splines and show that the smoothing spline is a sample function of a certain linear least-squares estimate. Estimation techniques are then used to derive a recursive algorithm for spline smoothing.

关键词： least-squares estimation recursive algorithms reproducing kernels Splien smoothing

来源：评论

学校读者我要写书评

暂无评论

Q-learning algorithms with random truncation bounds and applications to effective parallel computing

引用

JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS 2008年第2期137卷 435-451页

作者： Yin, G. Xu, C. Z. Wang, L. Y. Wayne State Univ Dept Math Detroit MI 48202 USA Wayne State Univ Dept Elect & Comp Engn Detroit MI 48202 USA

Motivated by an important problem of load balancing in parallel computing, this paper examines a modified algorithm to enhance Q-learning methods, especially in asynchronous recursive procedures for self-adaptive load distribution at run-time. Unlike the existing projection method that utilizes a fixed region, our algorithm employs a sequence of growing truncation bounds to ensure the boundedness of the iterates. Convergence and rates of convergence of the proposed algorithm are established. This class of algorithms has broad applications in signal processing, learning, financial engineering, and other related fields.

关键词： Q-learning recursive algorithms convergence rate of convergence

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：