Plan EL, Maloney A, Mentré F, Karlsson MO, Bertrand J (2012) Performance comparison of various maximum likelihood nonlinear mixed-effects estimation methods for dose-response models. J Pharmacokinet Pharmacodyn 31(4):299–320 Jönsson S, Kjellsson M, Karlsson M (2004) Estimating bias in population parameters for some models for repeated measures ordinal data using NONMEM and NLMIXED. Savic R, Lavielle M (2009) Performance in population models for count data, part II: a new SAEM algorithm. Plan E, Maloney A, Trocóniz I, Karlsson M (2009) Performance in population models for count data, part I: maximum likelihood approximations. AAPS J 9(1):E60–E83īeal S, Sheiner L, Boeckmann A, Bauer R (2009) “NONMEM user’s guides (1989–2009)” Techical Report, Icon Development Solutions, Ellicott City
#CURRENT NONMEM VERSION SOFTWARE#
765 in The International Series in Engineering and Computer Science, Springer, New York, p 135–163īauer RJ, Guzy S, Ng C (2007) A survey of population analysis methods and software for complex pharmacokinetic and pharmacodynamic models with examples. In: Advanced methods of pharmacokinetic and pharmacodynamic systems analysis, vol 3, no. Comput Stat Data Anal 49(4):1020–1038īauer RJ, Guzy S (2004) Monte Carlo parametric expectation maximization (MC-PEM) method for analyzing population pharmacokinetic/pharmacodynamic data.
Kuhn E, Lavielle M (2005) Maximum likelihood estimation in nonlinear mixed effects models. The Bayesian Markov Chain Monte Carlo method, used in this study for point estimation, performed worst in all tested metrics. FOCE/LAPLACE was the method with the shortest runtime for all models, followed by iterative two-stage. The methods relative robustness differed between models and no method showed clear superior performance.
The method giving the lowest bias and highest precision across models was importance sampling, closely followed by FOCE/LAPLACE and stochastic approximation expectation-maximization. Average estimation time for each algorithm and each model was calculated from the runtimes reported by NM7. Simulations of 100 data sets were used to investigate robustness by comparing final estimates obtained after estimations starting from the true parameter values and initial estimates randomly generated using the CHAIN feature in NM7. Simulations of 500 data sets from each PD model were reanalyzed with the available estimation methods to investigate bias and precision. In this study, performance of the estimation methods available in NM7 was investigated with respect to bias, precision, robustness and runtime for a diverse set of PD models. The latest version, NONMEM 7 (NM7), includes several sampling-based estimation methods in addition to the classical methods. NONMEM is the most widely used software for population pharmacokinetic (PK)-pharmacodynamic (PD) analyses.