SEQUENTIAL EXPERIMENTAL-DESIGN FOR PRECISE PARAMETER-ESTIMATION .1. USE OF REPARAMETERIZATION
SEQUENTIAL EXPERIMENTAL-DESIGN FOR PRECISE PARAMETER-ESTIMATION .1. USE OF REPARAMETERIZATION
复制标题
DOI：
10.1021/i200028a034
复制
发表时间：
1985-01-01
期刊：
影响因子：
--
通讯作者：
BRISK, ML
中科院分区：
文献类型：
--
作者：
AGARWAL, AK;BRISK, ML
考虑yu=/(XU， K)+ eu(4)给出的速率模型，其中yu为测量的反应速率；徐= (xlu, x2u,…， xmu)为实验运行u的m个操作变量，u= 1,2 .....n;K=(ku k2，…， kp) T是参数。若随机误差eu， u= 1,2 .....n为独立正态分布，方差2为常数，则最小二乘参数估计K的方差协方差矩阵为V=
Design Criterion for Parameter Estimation Consider a rate model given by yu=/(XU, K)+ eu(4) where yu is the measured reaction rate; Xu=(xlu, x2u,..., xmu) are the m operating variables for experimental run u, u= 1, 2..... n; and K=(ku k2,..., kp) T are the pparameters. If the random errors eu, u= 1, 2..... n are independently and normally distributed with constant var-iance 2, then the variance-covariance matrix of the least square parameter estimates K is V=