Qualitative and Quantitative Experiment Design for Nonlinear Models

Abstract Designing an experiment for parameter estimation involves two steps. The first one is qualitative, and consists in selecting a suitable configuration of the input/output ports so as to make, if possible, all the parameters of interest identifiable. The second step is quantitative, and based on the optimization of a suitable criterion (with respect to the input shapes, sampling schedule,…) so as to get the maximum information from the data to be collected. When the model is nonlinear in the parameters, both steps present specific difficulties which are discussed in this paper. The practical importance of qualitative experiment design is illustrated by a very simple biological model. Various policies presented in the literature for quantitative experiment design are reviewed. Special emphasis is given to methods allowing uncertainty on the prior information to be taken into account.

[1]  Philip E. Gill,et al.  Numerical methods for constrained optimization , 1974 .

[2]  Eric Walter,et al.  Estimation of non-uniquely identifiable parameters via exhaustive modeling and membership set theory , 1986 .

[3]  Y. Lecourtier,et al.  THE TESTING OF STRUCTURAL PROPERTIES THROUGH SYMBOLIC COMPUTATION , 1987 .

[4]  P. Eykhoff,et al.  Input signal design for system identification: a comparative analysis , 1985 .

[5]  Corwin L. Atwood,et al.  Optimal and Efficient Designs of Experiments , 1969 .

[6]  S. Silvey,et al.  A geometric approach to optimal design theory , 1973 .

[7]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

[8]  W. J. Studden,et al.  Theory Of Optimal Experiments , 1972 .

[9]  Eric Walter,et al.  A general-purpose global optimizer: implementation and applications , 1984 .

[10]  Gerhard Emig,et al.  Sequential experimental design procedures for precise parameter estimation in ordinary differential equations , 1975 .

[11]  Martin B. Zarrop,et al.  Optimal experiment design for dynamic system identification , 1977 .

[12]  W. G. Hunter,et al.  The use of prior distributions in the design of experiments for parameter estimation in non-linear situations: multiresponse case. , 1967, Biometrika.

[13]  Jürgen Pilz,et al.  Once more: optimal experimental design for regression models (with discussion) , 1987 .

[14]  L. Endrenyi.  Design of Experiments for Estimating Enzyme and Pharmacokinetic Parameters , 1981 .

[15]  J J DiStefano,et al.  Tracer experiment design for unique identification of nonlinear physiological systems. , 1976, The American journal of physiology.

[16]  R. Bellman,et al.  On structural identifiability , 1970 .

[17]  Changbao Wu,et al.  Some Algorithmic Aspects of the Theory of Optimal Designs , 1978 .

[18]  A. Dvoretzky.  On Stochastic Approximation , 1956 .

[19]  Joseph J. DiStefano,et al.  Algorithms, software and sequential optimal sampling schedule designs for pharmacokinetic and physiologic experiments , 1982 .

[20]  W. G. Hunter,et al.  The Experimental Study of Physical Mechanisms , 1965 .

[21]  Norbert Gaffke,et al.  Further Characterizations of Design Optimality and Admissibility for Partial Parameter Estimation in Linear Regression , 1987 .

[22]  Yves Lecourtier,et al.  Identifiability and distinguishability testing via computer algebra , 1985 .

[23]  L Pronzato,et al.  Optimal experiment design for nonlinear models subject to large prior uncertainties. , 1987, The American journal of physiology.

[24]  Harold W. Sorenson,et al.  Parameter estimation: Principles and problems , 1980 .

[25]  J. Kiefer.  General Equivalence Theory for Optimum Designs (Approximate Theory) , 1974 .

[26]  Alessandra Giovagnoli,et al.  Bayes D-optimal and E-optimal block designs , 1983 .

[27]  Claudio Cobelli,et al.  Identifiability results on some constrained compartmental systems , 1979 .

[28]  H. Chernoff.  Locally Optimal Designs for Estimating Parameters , 1953 .

[29]  J. Jacquez,et al.  Numerical parameter identifiability and estimability: Integrating identifiability, estimability, and optimal sampling design , 1985 .

[30]  G. Saridis.  Stochastic approximation methods for identification and control--A survey , 1974 .

[31]  H. Wynn.  Results in the Theory and Construction of D‐Optimum Experimental Designs , 1972 .

[32]  Anthony C. Atkinson,et al.  The Design of Experiments for Parameter Estimation , 1968 .

[33]  S. Audoly,et al.  On the identifiability of linear compartmental systems: a revisited transfer function approach based on topological properties , 1983 .

[34]  E. H. Twizell.  The mathematical modeling of metabolic and endocrine systems: E.R. Carson, C. Cobelli and L. Finkelstein John Wiley and Sons, Chichester, Sussex, UK, 394 pp., £45.15, 1983 , 1984 .

[35]  Y. Lecourtier,et al.  Comments on "On parameter and structural identifiablility: Nonunique observability/reconstructibility for identifiable systems, other ambiguities, and new definitions" , 1981 .

[36]  R. Schoenfeld,et al.  Invariants in Experimental Data on Linear Kinetics and the Formulation of Models , 1956 .

[37]  J. Kiefer,et al.  The Equivalence of Two Extremum Problems , 1960, Canadian Journal of Mathematics.

[38]  Graham C. Goodwin,et al.  Coupled design of test signals, sampling intervals, and filters for system identification , 1974 .

[39]  W. J. Studden,et al.  Optimal Experimental Designs , 1966 .

[40]  J. Kiefer,et al.  Optimum Designs in Regression Problems , 1959 .

[41]  H. L. Lucas,et al.  DESIGN OF EXPERIMENTS IN NON-LINEAR SITUATIONS , 1959 .

[42]  A. Mallet.  A maximum likelihood estimation method for random coefficient regression models , 1986 .

[43]  E. Walter,et al.  Identifiability and distinguishability of fundamental parameters in catalytic methanation , 1986 .

[44]  K. Chaloner.  Optimal Bayesian Experimental Design for Linear Models , 1984 .

[45]  L. White.  An extension of the General Equivalence Theorem to nonlinear models , 1973 .

[46]  R. Brooks.  On the choice of an experiment for prediction in linear regression , 1974 .

[47]  Identifiability: Its Role in Design of Pharmacokinetic Experi ments , 1982, IEEE Transactions on Biomedical Engineering.

[48]  J J DiStefano,et al.  Optimized blood sampling protocols and sequential design of kinetic experiments. , 1981, The American journal of physiology.

[49]  J F Boisvieux,et al.  Alternative approaches to estimation of population pharmacokinetic parameters: comparison with the nonlinear mixed-effect model. , 1984, Drug metabolism reviews.

[50]  David H. Anderson.  Structural properties of compartmental models , 1982 .

[51]  Robert E. Kalaba,et al.  Optimal inputs for blood glucose regulation parameter estimation, part 3 , 1981 .

[52]  W. G. Hunter,et al.  The use of prior distributions in the design of experiments for parameter estimation in non-linear situations. , 1967, Biometrika.

[53]  E. Walter,et al.  How to Design Experiments that are Robust to Parameter Uncertainty , 1985 .

[54]  W. G. Hunter,et al.  Design of experiments for parameter estimation in multiresponse situations , 1966 .

[55]  Changbao Wu,et al.  Asymptotic inference from sequential design in a nonlinear situation , 1985 .

[56]  Claudio Cobelli,et al.  Optimal input design for identification of compartmental models: theory and application to a model of glucose kinetics , 1985 .

[57]  Andrej Pázman,et al.  Foundations of Optimum Experimental Design , 1986 .

[58]  D. W. Behnken.  Estimation of copolymer reactivity ratios: An example of nonlinear estimation , 1964 .

[59]  Gerald M. Saidel,et al.  Analysis of iron kinetics: Identifiability, experiment design, and deterministic interpretations of a stochastic model , 1984 .

[60]  E. Walter,et al.  Robust experiment design via stochastic approximation , 1985 .

[61]  L. Balant.  Applicability of different types of models in health and disease. , 1984, Drug metabolism reviews.

[62]  Raman K. Mehra,et al.  Optimal input signals for parameter estimation in dynamic systems--Survey and new results , 1974 .

[63]  G. Goodwin,et al.  On optimal choice of sampling strategies for linear system identification , 1976 .

[64]  Eric Walter,et al.  On the structural output distinguishability of parametric models, and its relations with structural identifiability , 1984 .

[65]  H. Pohjanpalo.  System identifiability based on the power series expansion of the solution , 1978 .

[66]  E. Aiyoshi,et al.  Necessary conditions for min-max problems and algorithms by a relaxation procedure , 1980 .

[67]  I. W. Saunders,et al.  A model for myxomatosis. , 1980 .

[68]  E. Walter,et al.  Robust nonlinear parameter estimation in the bounded noise case , 1986, 1986 25th IEEE Conference on Decision and Control.

[69]  R. J. Brooks.  Optimal regression design for control in linear regression , 1977 .

[70]  D. Titterington,et al.  Inference and sequential design , 1985 .

[71]  Lennart Ljung,et al.  Optimal experiment designs with respect to the intended model application , 1986, Autom..

[72]  J. DiStefano,et al.  Optimal nonuniform sampling interval and test-input design for identification of physiological systems from very limited data , 1979 .

[73]  K. Spingarn,et al.  Optimal Inputs for Nonlinear Process Parameter Estimation , 1974, IEEE Transactions on Aerospace and Electronic Systems.

[74]  M. Milanese,et al.  Structural identifiability of compartmental models and pathophysiological information from the kinetics of drugs , 1975 .

[75]  Claudio Cobelli,et al.  Identifiability of compartmental systems and related structural properties , 1979 .

[76]  Claudio Cobelli,et al.  Minimal sampling schedules for identification of dynamic models of metabolic systems of clinical interest: case studies for two liver function tests , 1983 .

[77]  Elliot M. Landaw,et al.  Optimal multicompartmental sampling designs for parameter estimation: Practical aspects of the identification problem , 1982 .

[78]  R. C. St. John,et al.  D-Optimality for Regression Designs: A Review , 1975 .

[79]  Viatcheslav B. Melas.  Optimal designs for exponential regression , 1978 .

[80]  Valerii V. Fedorov,et al.  Duality of optimal designs for model discrimination and parameter estimation , 1986 .

[81]  H. Wynn,et al.  The Convergence of General Step-Length Algorithms for Regular Optimum Design Criteria , 1978 .

[82]  T. Brubaker,et al.  Nonlinear Parameter Estimation , 1979 .

[83]  Friedrich Pukelsheim,et al.  General Differential and Lagrangian Theory for Optimal Experimental Design , 1983 .

[84]  DiStefano Jj rd.  Design and optimization of tracer experiments in physiology and medicine. , 1980 .

[85]  Sensitizing Inputs and Optimal Sampling Instants , 1979 .

[86]  Z. Galil,et al.  Comparison of Simplex Designs for Quadratic Mixture Models , 1977 .

[87]  S. Vajda,et al.  IDENTIFIABILITY OF POLYNOMIAL SYSTEMS: STRUCTURAL AND NUMERICAL ASPECTS , 1987 .

[88]  Peter D. H. Hill,et al.  A Review of Experimental Design Procedures for Regression Model Discrimination , 1978 .

[89]  V. Fedorov,et al.  Convex design theory 1 , 1980 .

[90]  J. Eisenfeld.  New techniques for structural identifiability for large linear and nonlinear compartmental systems , 1982 .

[91]  H. M. Tsuchiya,et al.  On the accuracy of determining rate constants in enzymatic reactions , 1967 .

[92]  David B. Johnson,et al.  Efficient biokinetic experimental designs , 1975 .

[93]  M. J. Box,et al.  Estimation and Design Criteria for Multiresponse Non‐Linear Models with Non‐Homogeneous Variance , 1972 .

[94]  M. Rowland,et al.  Absorption kinetics of aspirin in man following oral administration of an aqueous solution. , 1972, Journal of pharmaceutical sciences.

[95]  M. J. Box.  Some Experiences with a Nonlinear Experimental Design Criterion , 1970 .

[96]  Isabella Verdinelli,et al.  Computing Bayes D‐ and A‐Optimal Block Designs for a Two‐Way Model , 1983 .

[97]  Lennart Ljung,et al.  Benefits of Feedback in Experiment Design , 1985 .

[98]  John F. MacGregor,et al.  The analysis and design of binary vapour‐liquid equilibrium experiments. Part II: The design of experiments , 1977 .

[99]  Park M. Reilly,et al.  Guidelines for the optimal design of experiment to estimate parameters in first order kinetic models , 1977 .

[100]  C. Atwood.  Sequences Converging to $D$-Optimal Designs of Experiments , 1973 .