References of "Lambert, Philippe"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailSimulation-based study comparing multiple imputation methods for non-monotone missing ordinal data in longitudinal settings
Donneau, Anne-Françoise ULg; Mauer, Murielle; Lambert, Philippe ULg et al

in Journal of Biopharmaceutical Statistics (in press)

The application of multiple imputation (MI) techniques as a preliminary step to handle missing values in data analysis is well established. The MI method can be classified into two broad classes, the ... [more ▼]

The application of multiple imputation (MI) techniques as a preliminary step to handle missing values in data analysis is well established. The MI method can be classified into two broad classes, the joint modeling and the fully conditional specification approaches. Their relative performance for the longitudinal ordinal data setting under the missing at random (MAR) assumption is not well documented. This paper intends to fill this gap by conducting a large simulation study on the estimation of the parameters of a longitudinal proportional odds model. The two MI methods are also illustrated in quality of life data from a cancer clinical trial. [less ▲]

Detailed reference viewed: 28 (9 ULg)
Full Text
See detailSemiparametric Bayesian frailty model for clustered interval-censored data
Cetinyürek, Aysun ULg; Lambert, Philippe ULg

Report (2014)

The shared frailty model is one of the popular tool to analyze correlated right-censored time-to-event data. In the shared frailty model, the latent frailty is assumed to be shared by the members of a ... [more ▼]

The shared frailty model is one of the popular tool to analyze correlated right-censored time-to-event data. In the shared frailty model, the latent frailty is assumed to be shared by the members of a cluster and is assigned a parametric distribution, typically, a gamma distribution due to its conjugacy. However, in case of interval-censored time-to-event data, the inclusion of gamma frailties results in complicated intractable likelihoods, where the conjugacy property does not hold anymore. Here, we propose a semiparametric Bayesian frailty model for analyzing such data. We discuss three parametric specifications for frailty distribution in the analysis of interval-censored data. Afterwards we call particular attention to nonparametric specification of the frailty distribution. The results of the simulation study suggest that the proposed approach is robust to misspecification of the frailty distribution. Moreover, the performance of the proposed methodology is quite good in practical situations where the frailty distribution is multimodal or skewed. The approach is applied to dental data arising from the Signal Tandmobiel Study. [less ▲]

Detailed reference viewed: 26 (7 ULg)
Full Text
See detailEstimation and approximation in multidimensional dynamics
Frasso, Gianluca ULg; Jaeger, Jonathan ULg; Lambert, Philippe ULg

E-print/Working paper (2013)

Differential equations (DEs) are commonly used to describe dynamic systems evolving in one (ordinary differential equations or ODEs) or in more than one dimensions (partial differential equations or PDEs ... [more ▼]

Differential equations (DEs) are commonly used to describe dynamic systems evolving in one (ordinary differential equations or ODEs) or in more than one dimensions (partial differential equations or PDEs). In real data applications the parameters involved in the DE models are usually unknown and need to be estimated from the available measurements together with the state function. In this paper, we present frequentist and Bayesian approaches for the joint estimation of the parameters and of the state functions involved in PDEs. We also propose two strategies to include differential (initial and/or boundary) conditions in the estimation procedure. We evaluate the performances of the proposed strategy on simulated and real data applications. [less ▲]

Detailed reference viewed: 34 (7 ULg)
Full Text
Peer Reviewed
See detailBayesian P-spline estimation in hierarchical models specified by systems of affine differential equations
Jaeger, Jonathan ULg; Lambert, Philippe ULg

in Statistical Modelling : An International Journal (2013), 13

Ordinary differential equations (ODEs) are widely used to model physical, chemical and biological processes. Current methods for parameter estimation can be computationally intensive and/or not suitable ... [more ▼]

Ordinary differential equations (ODEs) are widely used to model physical, chemical and biological processes. Current methods for parameter estimation can be computationally intensive and/or not suitable for inference and prediction. Frequentist approaches based on ODE-penalized smoothing techniques have recently solved part of these drawbacks. A full Bayesian approach based on ODE-penalized B-splines is proposed to jointly estimate ODE parameters and state functions from affine systems of differential equations. Simulations inspired by pharmacokinetic studies show that the proposed method provides comparable results to methods based on explicit solution of the ODEs and outperforms the frequentist ODE-penalized smoothing approach. The basic model is extended to a hierarchical one in order to study cases where several subjects are involved. This Bayesian hierarchical approach is illustrated on real data for the study of perfusion ratio after a femoral artery occlusion. Model selection is feasible through the analysis of the posterior distributions of the ODE adhesion parameters and is illustrated on a real pharmacokinetic dataset. [less ▲]

Detailed reference viewed: 31 (9 ULg)
Full Text
Peer Reviewed
See detailA Bayesian Design Space for analytical methods based on multivariate models and predictions
Lebrun, Pierre ULg; Boulanger, Bruno ULg; Debrus, Benjamin ULg et al

in Journal of Biopharmaceutical Statistics (2013), 23

The International Conference for Harmonization (ICH) has released regulatory guidelines for Pharmaceutical Development. In the document ICH Q8, The Design Space of a process is presented as the set of ... [more ▼]

The International Conference for Harmonization (ICH) has released regulatory guidelines for Pharmaceutical Development. In the document ICH Q8, The Design Space of a process is presented as the set of factor settings providing satisfactory results. However, ICH Q8 does not propose any practical methodology to define, derive and compute Design Space. In parallel, in the last decades, it has been observed that the diversity and the quality of analytical methods have evolved exponentially allowing substantial gains in selectivity and sensitivity. However, there is still a lack for a rationale towards the development of robust separation methods in a systematic way. Applying ICH Q8 to analytical methods provides a methodology for predicting a region of the space of factors in which results will be reliable. Combining design of experiments and Bayesian standard multivariate regression, an identified form of the predictive distribution of a new response vector has been identified and used, under non-informative as well as informative prior distributions of the parameters. From the responses and their predictive distribution, various critical quality attributes can be easily derived. This Bayesian framework was then extended to the multi-criteria setting to estimate the predictive probability that several critical quality attributes will be jointly achieved in the future use of an analytical method. An example based on a high-performance liquid chromatography (HPLC) method is given. For this example, a constrained sampling scheme was applied to ensure the modeled responses have desirable properties. [less ▲]

Detailed reference viewed: 177 (58 ULg)
Full Text
Peer Reviewed
See detailNonparametric additive location-scale models for interval censored data
Lambert, Philippe ULg

in Statistics and computing (2013), 23

An nonparametric additive model for the location and dispersion of a continuous response with an arbitrary smooth conditional distribution is proposed. B-splines are used to specify the three components ... [more ▼]

An nonparametric additive model for the location and dispersion of a continuous response with an arbitrary smooth conditional distribution is proposed. B-splines are used to specify the three components of the model. It can deal with interval censored data and multiple covariates. After a simulation study, the relation between age, the number of years of full-time education and the net income (provided as intervals) available per person in Belgian households is studied from survey data. [less ▲]

Detailed reference viewed: 45 (11 ULg)
Full Text
See detailBayesian ODE-penalized B-spline model with Gaussian mixture as error distribution
Jaeger, Jonathan ULg; Lambert, Philippe ULg

Scientific conference (2012, July 18)

In the standard Bayesian ODE-penalized B-spline approach, it is assumed that the error distribution is homogeneous Gaussian. But, in many applications, the normal assumption for the error distribution is ... [more ▼]

In the standard Bayesian ODE-penalized B-spline approach, it is assumed that the error distribution is homogeneous Gaussian. But, in many applications, the normal assumption for the error distribution is not a realistic choice. The goal of this paper is to extend the standard Bayesian ODE-penalized B-spline approach to settings where the error term distribution can be described using a mixture of normals. [less ▲]

Detailed reference viewed: 19 (2 ULg)
Full Text
See detailBayesian penalized smoothing approaches in models specified using affine differential equations with unknown error distributions
Jaeger, Jonathan ULg; Lambert, Philippe ULg

E-print/Working paper (2012)

A full Bayesian approach based on ODE-penalized B-splines and penalized Gaussian mixture is proposed to jointly estimate ODE-parameters, state function and error distribution from the observation of some ... [more ▼]

A full Bayesian approach based on ODE-penalized B-splines and penalized Gaussian mixture is proposed to jointly estimate ODE-parameters, state function and error distribution from the observation of some state functions involved in systems of affine differential equations. Simulations inspired by pharmacokinetic studies show that the proposed method provides comparable results to the method based on the standard ODE-penalized B-spline approach (i.e. with the Gaussian error distribution assumption) and outperforms the standard ODE-penalized B-splines when the distribution is not Gaussian. This methodology is illustrated on the Theophylline dataset. [less ▲]

Detailed reference viewed: 15 (3 ULg)
Full Text
See detailOn the use of adhesion parameters to validate models specified using systems of affine differential equations
Jaeger, Jonathan ULg; Lambert, Philippe ULg

E-print/Working paper (2012)

A strategy for the selection of system of differential equations is proposed based on Bayesian ODE-penalized B-spline approach. It estimates the ODE parameters, approximates the solution of the ODE model ... [more ▼]

A strategy for the selection of system of differential equations is proposed based on Bayesian ODE-penalized B-spline approach. It estimates the ODE parameters, approximates the solution of the ODE model and quantifies the suitability of the proposed differential equations to model the dynamics of the observed state functions. Simulation study confirms that these ODE-adhesion parameters are able to question a system of differential equations as a descriptor of the dynamics in the state functions. This methodology is illustrated on a pharmacokinetic study. [less ▲]

Detailed reference viewed: 14 (1 ULg)
Full Text
Peer Reviewed
See detailTesting conditional asymmetry: a residual-based approach
Lambert, Philippe ULg; Laurent, Sébastien; Veredas, Davis

in Journal of Economic Dynamics & Control (2012), 36

We propose three residual-based tests for conditional asymmetry. The distribution is assumed to fall into the class of skewed distributions of Fernandez and Steel (1998). In this class, asymmetry is ... [more ▼]

We propose three residual-based tests for conditional asymmetry. The distribution is assumed to fall into the class of skewed distributions of Fernandez and Steel (1998). In this class, asymmetry is measured by the ratio between the probabilities of being larger and smaller than the mode. Estimation is performed under the null hypothesis of constant asymmetry of the innovations and, in a second step, tests for conditional asymmetry are performed on generalized residuals through parametric and non- parametric methods. We derive the asymptotic distribution of the tests that incorporates the uncertainty of the estimated parameters. A Monte Carlo study shows that neglecting this uncertainty severely biases the tests. An empirical application on a basket of daily returns reveals that financial data often present dynamics in the conditional skewness. [less ▲]

Detailed reference viewed: 9 (2 ULg)
Full Text
Peer Reviewed
See detailEfficacy of pre-ascent climbing route visual inspection in indoor sport climbing
Sanchez, Xavier; Lambert, Philippe ULg; Jones, G. et al

in Scandinavian Journal of Medicine & Science in Sports (2012), 22

Pre-ascent climbing route visual inspection (route preview) has been suggested as a key climbing performance para- meter although its role has never been verified experimentally. We examined the efficacy ... [more ▼]

Pre-ascent climbing route visual inspection (route preview) has been suggested as a key climbing performance para- meter although its role has never been verified experimentally. We examined the efficacy of this perceptual-cognitive skill on indoor sport climbing performance. Twenty-nine male climbers, divided into intermediate, advanced and expert climbing level groups, climbed two indoor sport routes matching their climbing level and, where applicable, routes below their climbing level. At each level, one route was climbed with a preview, where participants benefited from a 3-min pre-ascent climbing route visual inspection. Performance was assessed in terms of output (route comple- tion) and form (number and duration of moves and stops). Route preview did not influence the output performance. Climbers using visual inspection were no more likely to finish the ascent than those without the option of using visual inspection. Conversely, route preview did influence form performance; climbers made fewer, and shorter stops during their ascent following a preview of the route. Form performances differences remained when baseline ability levels were taken into account, although for shorter duration of stops only with expert climbers benefiting most from route preview. The ability to visually inspect a climb before its ascent may represent an essential component of performance optimization. [less ▲]

Detailed reference viewed: 32 (4 ULg)
Full Text
Peer Reviewed
See detailSmooth estimation of survival functions and hazard ratios from interval-censored data using Bayesian penalized B-splines
Cetinyürek, Aysun ULg; Lambert, Philippe ULg

in Statistics in Medicine (2011), 30(1), 75-90

We discuss the use of Bayesian P-spline and of the composite link model to estimate survival functions and hazard-ratios from interval-censored data. If one further assumes proportionality of the hazards ... [more ▼]

We discuss the use of Bayesian P-spline and of the composite link model to estimate survival functions and hazard-ratios from interval-censored data. If one further assumes proportionality of the hazards, the proposed strategy provides a smoothed estimate of the baseline hazard along with estimates of global covariate effects. The frequentist properties of our Bayesian estimators are assessed by an extensive simulation study. We further illustrate the methodology by two examples showing that the proportionality of the hazards might also be found inappropriate from interval-censored data. [less ▲]

Detailed reference viewed: 29 (12 ULg)
Full Text
Peer Reviewed
See detailComments on: Inference in multivariate Archimedean copulas models
Lambert, Philippe ULg

in Test (2011), 20

Detailed reference viewed: 16 (0 ULg)
Full Text
See detailHealth insurance coverage and adverse selection
Lambert, Philippe ULg; Perelman, Sergio ULg; Pestieau, Pierre ULg et al

in Börsch-Supan, Axel; Brandt, Martina; Hank, Karsten (Eds.) et al The Individual and the Welfare State: Life Histories in Europe (2011)

Detailed reference viewed: 39 (4 ULg)
Full Text
Peer Reviewed
See detailSmooth semiparametric and nonparametric Bayesian estimation of bivariate densities from bivariate histogram data
Lambert, Philippe ULg

in Computational Statistics & Data Analysis (2011), 55

Penalized B-splines combined with the composite link model are used to estimate a bivariate density from a histogram with wide bins. The goals are multiple: they include the visualization of the ... [more ▼]

Penalized B-splines combined with the composite link model are used to estimate a bivariate density from a histogram with wide bins. The goals are multiple: they include the visualization of the dependence between the two variates, but also the estimation of derived quantities like Kendall’s tau, conditional moments and quantiles. Two strategies are proposed: the first one is semiparametric with flexible margins modeled using B-splines and a parametric copula for the dependence structure; the second one is nonparametric and is based on Kronecker products of the marginal B-spline bases. Frequentist and Bayesian estimations are described. A large simulation study quantifies the performances of the two methods under different dependence structures and for varying strengths of dependence, sample sizes and amounts of grouping. It suggests that Schwarz’s BIC is a good tool for classifying the competing models. The density estimates are used to evaluate conditional quantiles in two applications in social and in medical sciences. [less ▲]

Detailed reference viewed: 27 (6 ULg)
Full Text
Peer Reviewed
See detailAdaptive Bayesian P-splines to estimate varying regression coefficients: application to receptor occupancy estimation
Jullion, Astrid; Lambert, Philippe ULg; Vandenhende, François

in JSM Proceedings, Statistical Computing Section. Alexandria, VA: American Statistical Association. (2009)

In many applications of linear regression models, the regression coefficients are not regarded as fixed but as varying with another covariate named the effect modifier. A useful extension of the linear ... [more ▼]

In many applications of linear regression models, the regression coefficients are not regarded as fixed but as varying with another covariate named the effect modifier. A useful extension of the linear regression models are then varying coefficient models. To link the regression coefficient with the effect modifier, several methods may be considered. Here, we propose to use Bayesian P-splines to relate in a smoothed way the regression coefficient with the effect modifier. We show that this method enables a large level of flexibility: if necessary, adaptive penalties can be introduced in the model (Jullion and Lambert 2007) and linear constraints on the relation between the regression coefficient and the effect modifier may easily be added. We provide an illustration of the proposed method in a PET study where we want to estimate the relation between the Receptor Occupancy and the drug concentration in the plasma. As we work in a Bayesian setting, credibility sets are easily obtained for receptor occupancy, which take into account the uncertainty appearing at all the different estimation steps. [less ▲]

Detailed reference viewed: 39 (13 ULg)
Full Text
Peer Reviewed
See detailPharmacokinetic parameters estimation using adaptive Bayesian P-splines models
Jullion, Astrid; Lambert, Philippe ULg; Beck, Benoit et al

in Pharmaceutical Statistics (2009), 8

In preclinical and clinical experiments, pharmacokinetic (PK) studies are designed to analyse the evolution of drug concentration in plasma over time i.e. the PK profile. Some PK parameters are estimated ... [more ▼]

In preclinical and clinical experiments, pharmacokinetic (PK) studies are designed to analyse the evolution of drug concentration in plasma over time i.e. the PK profile. Some PK parameters are estimated in order to summarize the complete drug’s kinetic profile: area under the curve (AUC), maximal concentration (Cmax), time at which the maximal concentration occurs (tmax) and half-life time (t1/2). Several methods have been proposed to estimate these PK parameters. A first method relies on interpolating between observed concentrations. The interpolation method is often chosen linear. This method is simple and fast. Another method relies on compartmental modelling. In this case, nonlinear methods are used to estimate parameters of a chosen compartmental model. This method provides generally good results. However, if the data are sparse and noisy, two difficulties can arise with this method. The first one is related to the choice of the suitable compartmental model given the small number of data available in preclinical experiment for instance. Second, nonlinear methods can fail to converge. Much work has been done recently to circumvent these problems (J. Pharmacokinet. Pharmacodyn. 2007; 34:229–249, Stat. Comput., to appear, Biometrical J., to appear, ESAIM P&S 2004; 8:115–131). In this paper, we propose a Bayesian nonparametric model based on P-splines. This method provides good PK parameters estimation, whatever be the number of available observations and the level of noise in the data. Simulations show that the proposed method provides better PK parameters estimations than the interpolation method, both in terms of bias and precision. The Bayesian nonparametric method provides also better AUC and t1/2 estimations than a correctly specified compartmental model, whereas this last method performs better in tmax and Cmax estimations. We extend the basic model to a hierarchical one that treats the case where we have concentrations from different subjects. We are then able to get individual PK parameter estimations. Finally, with Bayesian methods, we can get easily some uncertainty measures by obtaining credibility sets for each PK parameter. [less ▲]

Detailed reference viewed: 60 (8 ULg)
Full Text
Peer Reviewed
See detailBayesian density estimation from grouped continuous data
Lambert, Philippe ULg; Eilers, Paul H.C.

in Computational Statistics & Data Analysis (2009), 53

Grouped data occur frequently in practice, either because of limited resolution of instruments, or because data have been summarized in relatively wide bins. A combination of the composite link model with ... [more ▼]

Grouped data occur frequently in practice, either because of limited resolution of instruments, or because data have been summarized in relatively wide bins. A combination of the composite link model with roughness penalties is proposed to estimate smooth densities from such data in a Bayesian framework. A simulation study is used to evaluate the performances of the strategy in the estimation of a density, of its quantiles and rst moments. Two illustrations are presented: the rst one involves grouped data of lead concentrations in the blood and the second one the number of deaths due to tuberculosis in The Netherlands in wide age classes. [less ▲]

Detailed reference viewed: 29 (10 ULg)