Banks, H. T.; Dediu, Sava; Ernstberger, Stacey L.; Kappel, Franz Generalized sensitivities and optimal experimental design. (English) Zbl 1279.62089 J. Inverse Ill-Posed Probl. 18, No. 1, 25-83 (2010). Summary: We consider the problem of estimating a modeling parameter \(\theta \) using a weighted least squares criterion \(J_d(y, \theta) = \sum^n_{i=1} \frac{1}{\sigma(t_i)^2}(y(t_i) - f(t_i, \theta))^2\) for given data \(y\) by introducing an abstract framework involving generalized measurement procedures characterized by probability measures. We take an optimal design perspective, the general premise (illustrated via examples) being that in any data collected, the information content with respect to estimating \(\theta \) may vary considerably from one time measurement to another, and in this regard some measurements may be much more informative than others. We propose mathematical tools which can be used to collect data in an almost optimal way, by specifying the duration and distribution of time sampling in the measurements to be taken, consequently improving the accuracy (i.e., reducing the uncertainty in estimates) of the parameters to be estimated.We recall the concepts of traditional and generalized sensitivity functions and use these to develop a strategy to determine the “optimal” final time T for an experiment; this is based on the time evolution of the sensitivity functions and of the condition number of the Fisher information matrix. We illustrate the role of the sensitivity functions as tools in optimal design of experiments, in particular in finding “best” sampling distributions. Numerical examples are presented throughout to motivate and illustrate the ideas. Cited in 15 Documents MSC: 62G08 Nonparametric regression and quantile regression 62H99 Multivariate analysis 90C31 Sensitivity, stability, parametric optimization 65K10 Numerical optimization and variational techniques 93B51 Design techniques (robust design, computer-aided design, etc.) 62B10 Statistical aspects of information-theoretic topics Keywords:least squares inverse problems; sensitivity and generalized sensitivity functions; Fisher information matrix; design of experiments PDFBibTeX XMLCite \textit{H. T. Banks} et al., J. Inverse Ill-Posed Probl. 18, No. 1, 25--83 (2010; Zbl 1279.62089) Full Text: DOI References: [1] Adelman H. M., A.I.A.A. Journal 24 pp 823– (1986) [2] Bai P., Math. Biosci. Eng. 4 pp 373– (2007) · Zbl 1133.60341 · doi:10.3934/mbe.2007.4.373 [3] DOI: 10.1088/0266-5611/17/1/308 · Zbl 1054.35121 · doi:10.1088/0266-5611/17/1/308 [4] DOI: 10.1515/1569394053978515 · doi:10.1515/1569394053978515 [5] Banks H. T., Chapter 6 pp 129– (2003) [6] Banks H. T., Chapter 11 pp 249– [7] DOI: 10.1515/jiip.2007.038 · Zbl 1141.34049 · doi:10.1515/jiip.2007.038 [8] Banks H. T., Math. Biosci. Eng. 4 pp 403– (2007) · Zbl 1175.34073 · doi:10.3934/mbe.2007.4.403 [9] DOI: 10.1515/JIIP.2007.001 · Zbl 1124.34002 · doi:10.1515/JIIP.2007.001 [10] DOI: 10.1214/aoms/1177697731 · Zbl 0193.47201 · doi:10.1214/aoms/1177697731 [11] DOI: 10.1137/1101016 · doi:10.1137/1101016 [12] DOI: 10.1114/1.207 · doi:10.1114/1.207 This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.