Abstract
Complex biological processes are usually experimented along time among a collection of individuals, longitudinal data are then available. The statistical challenge is to better understand the underlying biological mechanisms. A standard statistical approach is mixed-effects model where the regression function is highly-developed to describe precisely the biological processes (solutions of multi-dimensional ordinary differential equations or of partial differential equation). A classical estimation method relies on coupling a stochastic version of the EM algorithm with a Monte Carlo Markov Chain algorithm. This algorithm requires many evaluations of the regression function. This is clearly prohibitive when the solution is numerically approximated with a time-consuming solver. In this paper a meta-model relying on a Gaussian process emulator is proposed to approximate the regression function, that leads to what is called a mixed meta-model. The uncertainty of the meta-model approximation can be incorporated in the model. A control on the distance between the maximum likelihood estimates of the mixed meta-model and the maximum likelihood estimates of the exact mixed model is guaranteed. Eventually, numerical simulations are performed to illustrate the efficiency of this approach.
Similar content being viewed by others
References
Äijö, T., Lähdesmäki, H.: Learning gene regulatory networks from gene expression measurements using non-parametric molecular kinetics. Bioinformatics 25(22), 2937–2944 (2009)
Aronszajn, N.: Theory of reproducing kernel. Trans. Am. Math. Soc. 68(3), 337–404 (1950)
Barbillon, P., Celeux, G., Grimaud, A., Lefebvre, Y., Rocquigny, E.D.: Nonlinear methods for inverse statistical problems. Comput. Stat. Data Anal. 55(1), 132–142 (2011)
Chatterjee, A., Guedj, J.: Mathematical modelling of HCV infection: what can it teach us in the era of direct-acting antiviral agents? Antivir. Ther. 17(6 Pt B), 1171–1182 (2012)
Davidian, M., Giltinan, D.: Nonlinear Models to Repeated Measurement Data. Chapman and Hall, London (1995)
Delyon, B., Lavielle, M., Moulines, E.: Convergence of a stochastic approximation version of the EM algorithm. Ann. Stat. 27, 94–128 (1999)
Dempster, A., Laird, N., Rubin, D.: Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B 39, 1–38 (1977)
Donnet, S., Samson, A.: Estimation of parameters in incomplete data models defined by dynamical systems. J. Stat. Plan. Inference 137, 2815–2831 (2007)
Fang, K., Li, R., Sudjianto, A.: Design and Modeling for Computer Experiments (Computer Science & Data Analysis). Chapman & Hall/CRC, Boca Raton (2005)
Fu, S., Celeux, G., Bousquet, N., Couplet, M.: Bayesian inference for inverse problems occurring in uncertainty analysis. International Journal for Uncertainty Quantification 5(1), 73–98 (2014)
Grenier, E., Louvet, V., Vigneaux, P.: Parameter estimation in non-linear mixed effects models with SAEM Algorithm: extension from ODE to PDE. Math. Model. Numer. Anal. (ESAIM) 48(5), 1303 (2014)
Guedj, J., Thiébaut, R., Commenges, D.: Maximum likelihood estimation in dynamical models of HIV. Biometrics 63, 1198–2006 (2007)
Haario, H., Laine, M., Mira, A., Saksman, E.: Dram: efficient adaptive mcmc. Stat. Comput. 16(4), 339–354 (2006)
Johnson, M.E., Moore, L.M., Ylvisaker, D.: Minimax and maximin distance designs. J. Stat. Plan. Inference 26(2), 131–148 (1990)
Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Glob. Optim. 13(4), 455–492 (1998)
Kennedy, M., O’Hagan, A.: Bayesian calibration of computer models (with discussion). J. R. Stat. Soc. Ser. B. Methodol. 63(3), 425–464 (2001)
Kim, S., Li, L.: Statistical identifiability and convergence evaluation for nonlinear pharmacokinetic models with particle swarm optimization. Comput. Methods Progr. Biomed. 113(2), 413–432 (2014)
Koehler, J.R., Owen, A.B.: Computer experiments. Design and analysis of experiments, Handbook of Statistics, vol. 13, pp. 261–308. North-Holland, Amsterdam (1996)
Kuhn, E., Lavielle, M.: Maximum likelihood estimation in nonlinear mixed effects models. Comput. Stat. Data Anal. 49, 1020–1038 (2005)
Lavielle, M., Samson, A., Fermin, A., Mentre, F.: Maximum likelihood estimation of long term HIV dynamic models and antiviral response. Biometrics 67(1), 250–259 (2011)
Lophaven, N., Nielsen, H., Sondergaard, J.: DACE, a Matlab Kriging toolbox. Tech. Rep. IMM-TR-2002-12, DTU. http://www2.imm.dtu.dk/~hbn/dace/dace (2002)
Louis, T.A.: Finding the observed information matrix when using the EM algorithm. J. R. Stat. Soc. Ser. B Methodol. 44(2), 226–233 (1982)
Pinheiro, J., Bates, D.: Mixed-Effect Models in S and Splus. Springer, New York (2000)
Prasad, N., Rao, J.N.K.: The estimation of the mean squared error of small-area estimators. J. Am. Stat. Assoc. 85, 163–171 (1990)
Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning). MIT Press, Cambridge (2005)
Ribba, B., Kaloshi, G., Peyre, M., Ricard, D., Calvez, V., Tod, M., Cajavec-Bernard, B., Idbaih, A., Psimaras, D., Dainese, L., Pallud, J., Cartalat-Carel, S., Delattre, J., Honnorat, J., Grenier, E., Ducray, F.: A tumor growth inhibition model for low-grade glioma treated with chemotherapy or radiotherapy. Clin. Cancer Res. 18, 5071–5080 (2012)
Rougier, J.: Efficient emulators for multivariate deterministic functions. J. Comput. Graph. Stat. 17(4), 827–843 (2008)
Sacks, J., Schiller, S.B., Mitchell, T.J., Wynn, H.P.: Design and analysis of computer experiments. Stat. Sci. 4, 409–435 (1989)
Samson, A., Lavielle, M., Mentré, F.: The SAEM algorithm for group comparison tests in longitudinal data analysis based on non-linear mixed-effects model. Stat. Med. 26(27), 4860–4875 (2007)
Santner, T.J., Williams, B., Notz, W.: The Design and Analysis of Computer Experiments. Springer, New York (2003)
Schaback, R.: Error estimates and condition numbers for radial basis function interpolation. Adv. Comput. Math. 3(3), 251–264 (1995)
Schaback, R.: Kernel-based meshless methods. Tech. Rep., Institute for Numerical and Applied Mathematics, Georg-August-University Goettingen (2007)
Wei, G.C.G., Tanner, M.A.: Calculating the content and boundary of the highest posterior density region via data augmentation. Biometrika 77(3), 649–652 (1990)
Wolfinger, R.: Laplace’s approximation for nonlinear mixed models. Biometrika 80(4), 791–795 (1993)
Wu, H., Huang, Y., Acosta, E., Rosenkranz, S., Kuritzkes, D., Eron, J., Perelson, A., Gerber, J.: Modeling long-term HIV dynamics and antiretroviral response: effects of drug potency, pharmacokinetics, adherence, and drug resistance. J. Acquir. Immune Defic. Syndr. 39, 272–283 (2005)
Wu, Z.M., Schaback, R.: Local error estimates for radial basis function interpolation of scattered data. IMA J. Numer. Anal. 13, 13–27 (1992)
Acknowledgments
Adeline Samson has been supported by the LabEx PERSYVAL-Lab (ANR-11-LABX-0025-01). Les recherches menant aux présents résultats ont bénéficié d’un soutien financier du septiéme programme-cadre de l’Union européenne (7ePC/2007-2013) en vertu de la convention de subvention n 266638.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 Proof of Proposition 3
We have
Therefore, we start by studying \(|p(\mathbf {y}|\varvec{\psi };\varvec{\theta })-\tilde{p}_D(\mathbf {y}|\varvec{\psi };\varvec{\theta })|\):
Under the assumption that the functions f and \(m_D\) are uniformly bounded on the support of \(\psi \), there exists a constant \(C_y\) which is uniform according to \(\psi \) such that \(|2y_{ij}-f(t,\psi )-m_D(t,\psi )|\le C_y\). Proposition 1 implies that the approximation error due to the metamodel \(|f(t_{ij},\psi _i)-m_D(t_{ij},\psi _i)|\) is controlled by inequality (8):
Then there exists a constant \(C_y\) depending only on \(\mathbf {y}\) such that
Finally
\(\square \)
1.2 Proof of Proposition 4
We study the distance between the two likelihoods \(p_D\) and \(\tilde{p}_D\). As in Proposition 3, we start by studying \(|p_D(\mathbf {y}|\varvec{\psi };\varvec{\theta })-\tilde{p}_D(\mathbf {y}|\varvec{\psi };\varvec{\theta })|\). We consider two Gaussian distributions with same expectations and different covariance matrix. Thus this distance is maximum when \(\sum (y_{ij}-m_D(t_{ij},\psi _i))^2=0\). This yields
where we use that the determinant, as a product of eigen values, is smaller than a function of the trace of the matrix. Thus, the sum is over the diagonal of the matrix \(\mathbf {C}_D\) i.e. the sum of the variances. Then, we obtain that there exists a constant C such that
where the last inequality holds using Proposition 1. Finally, we obtain
The proof is similar for the distance between the two likelihoods \(\bar{p}_D\) and \(\tilde{p}_D\). \(\square \)
1.3 Details of assumptions for Proposition 2
- (M1) :
-
function f and the individual parameters distribution \(p(\psi ; \beta )\) are such that there exist functions \(g, \nu \) of \(\theta \) verifying
$$\begin{aligned} \log p(y, \psi ;\theta )= - g(\theta ) + \left\langle S(y,\psi ), \nu (\theta ) \right\rangle \end{aligned}$$where \(S(y, \psi )\) is a minimal sufficient statistic of the complete model, taking its value in a subset \({\mathcal {S}}\) and \( \left\langle \cdot , \cdot \right\rangle \) is the scalar product.
- (M2) :
-
The functions \(g(\theta )\) and \(\nu (\theta )\) are twice continuously differentiable on \(\varTheta \).
- (M3) :
-
The function \(\bar{s}:\varTheta \longrightarrow \mathcal {S}\) defined as \( \bar{s} (\theta ) = \int S(y,\psi ) p(\psi |y;\theta ) d\psi \) is continuously differentiable on \(\varTheta \).
- (M4) :
-
The function \(\ell (\theta ) = \log p(y,\theta )\) is continuously differentiable on \(\varTheta \) and
$$\begin{aligned} \partial _\theta \int p(y,\psi ;\theta ) d\psi = \int \partial _\theta p(y,\psi ;\theta ) d\psi . \end{aligned}$$ - (M5) :
-
Define \(L:\mathcal {S}\times \varTheta \longrightarrow \mathbb {R}\) as \( L(s,\theta ) = -g(\theta )+\langle s,\nu (\theta )\rangle \). There exists a function \(\hat{\theta }:\mathcal {S}\longrightarrow \varTheta \) such that
$$\begin{aligned} \forall \theta \in \varTheta , \quad \forall s \in \mathcal {S}, \quad L(s,\hat{\theta }(s))\ge L(s,\theta ). \end{aligned}$$ - (SAEM1) :
-
The positive decreasing sequence of the stochastic approximation \((\alpha _m)_{m \ge 0}\) is such that \(\sum _{m} \alpha _m = \infty \) and \(\sum _{m}\alpha ^2_m < \infty \).
- (SAEM2) :
-
\(\ell :\varTheta \rightarrow \mathbb {R}\) and \(\hat{\theta }:\mathcal {S}\rightarrow \varTheta \) are d times differentiable, where d is the dimension of \(S(y,\psi )\).
- (SAEM3) :
-
For all \(\theta \in \varTheta \), \(\int || S(y,\psi )||^2\, p(\psi |y;\theta ) d\psi < \infty \) and the function \(\varGamma (\theta )=Cov_\theta (S(\psi ))\) is continuous.
- (SAEM4) :
-
S is a bounded function.
- (SAEM5’) :
-
Let us denote \(\Pi _\theta \) the transition kernel of the PMCMC algorithm and \(\pi (\psi )=p(\psi |y;\theta )\) its stationary distribution. We assume that \(\Pi _\theta \) is Lipschitz in \(\theta \) and generates a ergodic chain such that for any starting point \((\varphi )\)
$$\begin{aligned} \sum _{m\ge 0} \Vert 1_{\varphi }\Pi _\theta ^m- \pi \Vert _{TV} <\infty \end{aligned}$$where \(\Vert \cdot \Vert _{TV}\) is the total variation of probability measures. This property is also called ergodicity of degree 2.
1.4 Additional results for Section 7
In Tables 1, 3 and 2, relative bias and relative root mean square error (RMSE) are displayed for each population parameter. The \(95\,\%\) coverage rates correspond to the coverage rate of the confidence interval on parameters based on the stochastic approximation of the Fisher Information matrix are also displayed in these tables. These results are obtained from 100 replications in Table 1 and from 1000 replications in Tables 3 and 2.
Rights and permissions
About this article
Cite this article
Barbillon, P., Barthélémy, C. & Samson, A. Parameter estimation of complex mixed models based on meta-model approach. Stat Comput 27, 1111–1128 (2017). https://doi.org/10.1007/s11222-016-9674-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-016-9674-x