Parameter Estimation and Variable Selection for Big Systems of Linear Ordinary Differential Equations: A Matrix-Based Approach.

Parameter Estimation and Variable Selection for Big Systems of Linear Ordinary Differential Equations: A Matrix-Based Approach.
复制标题

大线性常微分方程组的参数估计和变量选择:基于矩阵的方法

DOI:
10.1080/01621459.2017.1423074
复制
发表时间:
2019
影响因子:
3.7
通讯作者:
Wu H
Wu H
中科院分区:
数学1区
文献类型:
--
作者:
Wu L;Qiu X;Yuan YX;Wu H

文献摘要

参考文献

被引文献

相似文献

常微分方程被广泛用于复杂系统的动力学行为建模。由于需要在超高维的参数空间中进行非线性优化,因此,对具有线性常微分方程的“大系统”进行参数估计和变量选择是非常具有挑战性的。本文提出了一种基于相似变换和可分离最小二乘(SLS)思想的参数估计和变量选择方法。仿真研究表明,对于具有数千维、数百万参数的线性常微分方程系统,本文提出的基于矩阵的SLS方法能够更准确地估计系数矩阵,并能更好地进行变量选择,优于现有的直接最小二乘法和基于向量的两阶段方法.我们将该方法应用于两个真实的数据集:一个包含30个维度和930个未知参数的酵母细胞周期基因表达数据集和一个包含1250个维度和1,563,750个未知参数的标准普尔1500指数股票价格数据集,以说明所提出的大系统参数估计和变量选择方法的实用性和数值性能。
Ordinary differential equations (ODEs) are widely used to model the dynamic behavior of a complex system. Parameter estimation and variable selection for a “Big System” with linear ODEs are very challenging due to the need of nonlinear optimization in an ultra-high dimensional parameter space. In this article, we develop a parameter estimation and variable selection method based on the ideas of similarity transformation and separable least squares (SLS). Simulation studies demonstrate that the proposed matrix-based SLS method could be used to estimate the coefficient matrix more accurately and perform variable selection for a linear ODE system with thousands of dimensions and millions of parameters much better than the direct least squares (LS) method and the vector-based two-stage method that are currently available. We applied this new method to two real data sets: a yeast cell cycle gene expression data set with 30 dimensions and 930 unknown parameters and the Standard & Poor 1500 index stock price data with 1250 dimensions and 1,563,750 unknown parameters, to illustrate the utility and numerical performance of the proposed parameter estimation and variable selection method for big systems in practice.
DOI: 10.5705/ss.2012.304
发表时间: 2014-10
期刊: Statistica Sinica
影响因子: 1.4
作者:
Ding AA;Wu H
通讯作者: Wu H
DOI: 10.1186/gb-2006-7-2-r9
发表时间: 2006
期刊: Genome biology
影响因子: 12.3
作者:
Aragon AD;Quiñones GA;Thomas EV;Roy S;Werner-Washburne M
通讯作者: Werner-Washburne M
DOI: 10.1073/pnas.98.4.1693
发表时间: 2001-02-13
影响因子: 11.1
作者:
Holter, NS;Maritan, A;Banavar, JR
通讯作者: Banavar, JR
DOI: 10.1111/j.2517-6161.1995.tb02031.x
发表时间: 1995-01-01
影响因子: 5.8
作者:
BENJAMINI, Y;HOCHBERG, Y
通讯作者: HOCHBERG, Y
DOI: 10.1002/bimj.200900173
发表时间: 2010-08
影响因子: 1.7
作者:
Huang, Yangxin;Wu, Hulin;Acosta, Edward P.
通讯作者: Acosta, Edward P.