Density-based Monte Carlo filter and its applications in nonlinear stochastic differential equation models

Nonlinear stochastic differential equation models with unobservable state variables are now widely used in analysis of PK/PD data. Unobservable state variables are usually estimated with extended Kalman filter (EKF), and the unknown pharmacokinetic parameters are usually estimated by maximum likelihood estimator. However, EKF is inadequate for nonlinear PK/PD models, and MLE is known to be biased downwards. A density-based Monte Carlo filter (DMF) is proposed to estimate the unobservable state variables, and a simulation-based M estimator is proposed to estimate the unknown parameters in this paper, where a genetic algorithm is designed to search the optimal values of pharmacokinetic parameters. The performances of EKF and DMF are compared through simulations for discrete time and continuous time systems respectively, and it is found that the results based on DMF are more accurate than those given by EKF with respect to mean absolute error.

[1]  É. Pardoux,et al.  Stochastic Differential Equations , 2014 .

[2]  A. Gaetano,et al.  Stochastic vs. deterministic uptake of dodecanedioic acid by isolated rat livers , 2005, Bulletin of mathematical biology.

[3]  H. Madsen,et al.  Non-Linear Mixed-Effects Models with Stochastic Differential Equations: Implementation of an Estimation Algorithm , 2005, Journal of Pharmacokinetics and Pharmacodynamics.

[4]  Henrik Madsen,et al.  Stochastic Differential Equations in NONMEM®: Implementation, Application, and Comparison with Ordinary Differential Equations , 2005, Pharmaceutical Research.

[5]  A. V. D. Vaart,et al.  Efficient maximum likelihood estimation in semiparametric mixture models , 1996 .

[6]  H. Madsen,et al.  Using Stochastic Differential Equations for PK/PD Model Development , 2005, Journal of Pharmacokinetics and Pharmacodynamics.

[7]  Janine Mukuddem-Petersen,et al.  Continuous-time stochastic modelling of capital adequacy ratios for banks , 2006 .

[8]  Malene Højbjerre,et al.  A population-based Bayesian approach to the minimal model of glucose and insulin homeostasis. , 2005, Statistics in medicine.

[9]  Cristian Meza,et al.  REML Estimation of Variance Parameters in Nonlinear Mixed Effects Models Using the SAEM Algorithm , 2007, Biometrical journal. Biometrische Zeitschrift.

[10]  Umberto Picchini,et al.  Stochastic Differential Mixed‐Effects Models , 2010 .

[11]  Dan Simon,et al.  Optimal State Estimation: Kalman, H∞, and Nonlinear Approaches , 2006 .

[12]  Hisashi Tanizaki,et al.  Nonlinear Filters: Estimation and Applications , 1993 .

[13]  Hisashi Tanizaki,et al.  Nonlinear filters , 1993 .

[14]  Henrik Madsen,et al.  PKPD Model of Interleukin-21 Effects on Thermoregulation in Monkeys—Application and Evaluation of Stochastic Differential Equations , 2007, Pharmaceutical Research.

[15]  T. Hansen,et al.  Predictive performance for population models using stochastic differential equations applied on data from an oral glucose tolerance test , 2010, Journal of Pharmacokinetics and Pharmacodynamics.