arXiv subjects

Mark A. van de Wiel

Publications and source records attributed to Mark A. van de Wiel.

At least 19 recordsLinked to original sources

arXiv2026

Discovery and inference beyond linearity for epidemiological data by integrating Bayesian regression, tree ensembles and Shapley values

Machine Learning (ML) is gaining popularity in epidemiology and healthcare studies for hypothesis-free discovery of risk and protective factors. ML is strong at discovering nonlinearities and interactions, but this power is compromised by a lack of reliable inference. Although Shapley values provide local measures of features' effects, valid uncertainty quantification for these effects is typically lacking, thus precluding statistical inference. We propose RuleSHAP, a framework that addresses this limitation by combining a dedicated Bayesian sparse regression model with an improved tree-based rule generator and Shapley value attribution. RuleSHAP provides detection of nonlinear and interaction effects, with uncertainty quantification at the individual level as a key contribution. We derive an efficient formula for computing marginal Shapley values within this framework. We apply RuleSHAP to data from an epidemiological cohort to detect and infer several effects for high cholesterol and blood pressure, such as nonlinear interaction effects between features like age, sex, ethnicity, BMI and glucose level. To conclude, we demonstrate the validity of our framework on simulated data.

stat.ML↗

arXiv2026

Combined shrinkage of fixed and random effects in linear mixed models using empirical Bayes

A novel data-driven methodology is presented for the joint selection of prior parameters for both fixed and random effects in Linear Mixed Models (LMMs). This approach facilitates the estimation of complex random-effects structures, as well as potentially high-dimensional data. Although Bayesian frameworks require the specification of informative prior parameters, such values are often unavailable a priori - especially for random-effect covariances. The proposed method automates this selection through an Empirical Bayes framework, which maximizes the marginal likelihood using an efficient Laplace approximation. Numerical simulations demonstrate that this methodology significantly enhances parameter estimation accuracy and predictive performance. Finally, an application to a real-world air pollution and health dataset illustrates how the method enables the use of more sophisticated and statistically appropriate models to improve predictive outcomes.

stat.ME↗

arXiv2026

ProfileGLMM: a R Package Extending Bayesian Profile Regression using Generalised Linear Mixed Models

ProfileGLMM is an R package integrating Generalised Linear Mixed Models (GLMMs) as the outcome model for Bayesian profile regression. This statistical framework simultaneously i) explains the variation in the outcome and ii) clusters the observations based on a specified set of interdependent clustering covariates. The derived cluster memberships are then incorporated, alongside others, as explanatory variables in the regression to model the outcome. This framework efficiently handles complex, highly correlated covariate structures whose direct inclusion in a standard regression model would be statistically sub-optimal. ProfileGLMM significantly extends Bayesian profile regression's scope by resolving two key constraints of previous implementations: 1) it allows the analysis of hierarchical and longitudinal data structures through the inclusion of random effects, and 2) it enables the study of interactions between latent clusters and other observable covariates. ProfileGLMM accommodates various data types, supporting both continuous or binary outcomes and both categorical and continuous clustering covariates. Built on fast Rcpp code with minimal mandatory parameters, ProfileGLMM offers a flexible analytical tool. It significantly enhances the utility of profile regression for researchers in fields such as epidemiology, social sciences, and clinical studies dealing with complex data.

stat.ME↗

arXiv2026

Bivariate deconvolution for cancer detection after surgery

Detection of minimal residual disease (MRD) in cancer patients after surgery can provide an early marker for disease recurrence and guide subsequent treatment decisions. Accurate and sensitive estimation of tumour burden after cancer surgery may be obtained through liq- uid biopsies, measuring circulating tumour DNA (ctDNA) using, for example, mutation-based Variant Allele Frequency (VAF) values. However, to be applicable to all patients this ei- ther requires tumour-informed, patient-specific mutation panels or sensitive, tumour-agnostic genome-wide measurements. We propose a solution that accounts for patient-specific charac- teristics in genome-wide screens. For that, we introduce a bivariate deconvolution model to estimate tumour proportion from circulating cell-free DNA (cfDNA) methylation profiles of patients before and after surgery. The observations are modelled as a convolution of two bivariate latent variables, corresponding to tumour and background signals, mixed by the tumour proportion at each measurement. This bivariate approach links pre- and post-surgery measurements improving estimation of the tumour proportion after surgery, when the tumour signal is potentially very weak, or absent. We approximate likelihood of the convolution through a discretisation of the bivariate density for each latent variable into a two-dimensional grid for each pair of observations which allows for fast maximum likelihood estimation. We evaluate the predictive performance of the estimated post-surgery tumour proportions based on cfDNA methylation against available mutation-based VAF values in one-year recurrence-free survival.

stat.ME↗

arXiv2026

How important are the genes to explain the outcome - the asymmetric Shapley value as an honest importance metric for high-dimensional features

In clinical prediction settings the importance of a high-dimensional feature like genomics is often assessed by evaluating the change in predictive performance when adding it to a set of traditional clinical variables. This approach is questionable, because it does not account for collinearity nor known directionality of dependencies between variables. We suggest to use asymmetric Shapley values as a more suitable alternative to quantify feature importance in the context of a mixed-dimensional prediction model. We focus on a setting that is particularly relevant in clinical prediction: disease state as a mediating variable for genomic effects, with additional confounders for which the direction of effects may be unknown. We derive efficient algorithms to compute local and global asymmetric Shapley values for this setting. The former are shown to be very useful for inference, whereas the latter provide interpretation by decomposing any predictive performance metric into contributions of the features. Throughout, we illustrate our framework by a leading example: the prediction of progression-free survival for colorectal cancer patients.

stat.ML↗

arXiv2025

False Discovery estimation in Record Linkage

Integrating data from multiple sources expands research opportunities at low cost. However, due to different data collection processes and privacy constraints, unique identifiers are unavailable. Record Linkage (RL) algorithms address this by probabilistically linking records based on partially identifying variables. Since these variables lack the strength to perfectly combine information, RL procedures yield an imperfect set of linked records. Therefore, assessing the false discovery proportion (FDP) in RL is crucial for ensuring the reliability of subsequent analyses. In this paper, we introduce a novel method for estimating the FDP in RL for two overlapping data sets. We synthesise data from their estimated empirical distribution and use it along with real data in the linkage process. Since synthetic records cannot form links with real entities, they provide a means to estimate the amount of falsely linked pairs. Notably, this method applies to all RL techniques and across diverse settings where links and non-links have similar distributions -- typical in complex tasks with poorly discriminative linking variables and multiple records sharing similar information while representing different entities. By identifying the FDP in RL and selecting suitable model parameters, our approach enables to assess and improve the reliability of linked data. We evaluate its performance using established RL algorithms and benchmark data applications before deploying it to link siblings from the Netherlands Perinatal Registry, where the reliability of previous RL applications has never been confirmed. Through this application, we highlight the importance of accounting for linkage errors when studying mother-child dynamics in healthcare records.

stat.ME↗

arXiv2024

A Semi-supervised CART Model for Covariate Shift

Machine learning models used in medical applications often face challenges due to the covariate shift, which occurs when there are discrepancies between the distributions of training and target data. This can lead to decreased predictive accuracy, especially with unknown outcomes in the target data. This paper introduces a semi-supervised classification and regression tree (CART) that uses importance weighting to address these distribution discrepancies. Our method improves the predictive performance of the CART model by assigning greater weights to training samples that more accurately represent the target distribution, especially in cases of covariate shift without target outcomes. In addition to CART, we extend this weighted approach to generalized linear model trees and tree ensembles, creating a versatile framework for managing the covariate shift in complex datasets. Through simulation studies and applications to real-world medical data, we demonstrate significant improvements in predictive accuracy. These findings suggest that our weighted approach can enhance reliability in medical applications and other fields where the covariate shift poses challenges to model performance across various data distributions.

cs.LG↗

arXiv2024

Fusion of Tree-induced Regressions for Clinico-genomic Data

Cancer prognosis is often based on a set of omics covariates and a set of established clinical covariates such as age and tumor stage. Combining these two sets poses challenges. First, dimension difference: clinical covariates should be favored because they are low-dimensional and usually have stronger prognostic ability than high-dimensional omics covariates. Second, interactions: genetic profiles and their prognostic effects may vary across patient subpopulations. Last, redundancy: a (set of) gene(s) may encode similar prognostic information as a clinical covariate. To address these challenges, we combine regression trees, employing clinical covariates only, with a fusion-like penalized regression framework in the leaf nodes for the omics covariates. The fusion penalty controls the variability in genetic profiles across subpopulations. We prove that the shrinkage limit of the proposed method equals a benchmark model: a ridge regression with penalized omics covariates and unpenalized clinical covariates. Furthermore, the proposed method allows researchers to evaluate, for different subpopulations, whether the overall omics effect enhances prognosis compared to only employing clinical covariates. In an application to colorectal cancer prognosis based on established clinical covariates and 20,000+ gene expressions, we illustrate the features of our method.

stat.ME↗

arXiv2024

Co-data Learning for Bayesian Additive Regression Trees

Medical prediction applications often need to deal with small sample sizes compared to the number of covariates. Such data pose problems for prediction and variable selection, especially when the covariate-response relationship is complicated. To address these challenges, we propose to incorporate co-data, i.e. external information on the covariates, into Bayesian additive regression trees (BART), a sum-of-trees prediction model that utilizes priors on the tree parameters to prevent overfitting. To incorporate co-data, an empirical Bayes (EB) framework is developed that estimates, assisted by a co-data model, prior covariate weights in the BART model. The proposed method can handle multiple types of co-data simultaneously. Furthermore, the proposed EB framework enables the estimation of the other hyperparameters of BART as well, rendering an appealing alternative to cross-validation. We show that the method finds relevant covariates and that it improves prediction compared to default BART in simulations. If the covariate-response relationship is nonlinear, the method benefits from the flexibility of BART to outperform regression-based co-data learners. Finally, the use of co-data enhances prediction in an application to diffuse large B-cell lymphoma prognosis based on clinical covariates, gene mutations, DNA translocations, and DNA copy number data. Keywords: Bayesian additive regression trees; Empirical Bayes; Co-data; High-dimensional data; Omics; Prediction

stat.ML↗

arXiv2024

Guiding adaptive shrinkage by co-data to improve regression-based prediction and feature selection

The high dimensional nature of genomics data complicates feature selection, in particular in low sample size studies - not uncommon in clinical prediction settings. It is widely recognized that complementary data on the features, `co-data', may improve results. Examples are prior feature groups or p-values from a related study. Such co-data are ubiquitous in genomics settings due to the availability of public repositories. Yet, the uptake of learning methods that structurally use such co-data is limited. We review guided adaptive shrinkage methods: a class of regression-based learners that use co-data to adapt the shrinkage parameters, crucial for the performance of those learners. We discuss technical aspects, but also the applicability in terms of types of co-data that can be handled. This class of methods is contrasted with several others. In particular, group-adaptive shrinkage is compared with the better-known sparse group-lasso by evaluating feature selection. Finally, we demonstrate the versatility of the guided shrinkage methodology by showing how to `do-it-yourself': we integrate implementations of a co-data learner and the spike-and-slab prior for the purpose of improving feature selection in genetics studies.

stat.ME↗

arXiv2023

Linked shrinkage to improve estimation of interaction effects in regression models

We address a classical problem in statistics: adding two-way interaction terms to a regression model. As the covariate dimension increases quadratically, we develop an estimator that adapts well to this increase, while providing accurate estimates and appropriate inference. Existing strategies overcome the dimensionality problem by only allowing interactions between relevant main effects. Building on this philosophy, we implement a softer link between the two types of effects using a local shrinkage model. We empirically show that borrowing strength between the amount of shrinkage for main effects and their interactions can strongly improve estimation of the regression coefficients. Moreover, we evaluate the potential of the model for inference, which is notoriously hard for selection strategies. Large-scale cohort data are used to provide realistic illustrations and evaluations. Comparisons with other methods are provided. The evaluation of variable importance is not trivial in regression models with many interaction terms. Therefore, we derive a new analytical formula for the Shapley value, which enables rapid assessment of individual-specific variable importance scores and their uncertainties. Finally, while not targeting for prediction, we do show that our models can be very competitive to a more advanced machine learner, like random forest, even for fairly large sample sizes. The implementation of our method in RStan is fairly straightforward, allowing for adjustments to specific needs.

stat.ME↗

arXiv2023

Think before you shrink: Alternatives to default shrinkage methods can improve prediction accuracy, calibration and coverage

While shrinkage is essential in high-dimensional settings, its use for low-dimensional regression-based prediction has been debated. It reduces variance, often leading to improved prediction accuracy. However, it also inevitably introduces bias, which may harm two other measures of predictive performance: calibration and coverage of confidence intervals. Much of the criticism stems from the usage of standard shrinkage methods, such as lasso and ridge with a single, cross-validated penalty. Our aim is to show that readily available alternatives can strongly improve predictive performance, in terms of accuracy, calibration or coverage. For linear regression, we use small sample splits of a large, fairly typical epidemiological data set to illustrate this. We show that usage of differential ridge penalties for covariate groups may enhance prediction accuracy, while calibration and coverage benefit from additional shrinkage of the penalties. In the logistic setting, we apply an external simulation to demonstrate that local shrinkage improves calibration with respect to global shrinkage, while providing better prediction accuracy than other solutions, like Firth's correction. The benefits of the alternative shrinkage methods are easily accessible via example implementations using \texttt{mgcv} and \texttt{r-stan}, including the estimation of multiple penalties. A synthetic copy of the large data set is shared for reproducibility.

stat.ME↗

arXiv2022

Penalised regression with multiple sources of prior effects

In many high-dimensional prediction or classification tasks, complementary data on the features are available, e.g. prior biological knowledge on (epi)genetic markers. Here we consider tasks with numerical prior information that provide an insight into the importance (weight) and the direction (sign) of the feature effects, e.g. regression coefficients from previous studies. We propose an approach for integrating multiple sources of such prior information into penalised regression. If suitable co-data are available, this improves the predictive performance, as shown by simulation and application. The proposed method is implemented in the R package `transreg' (https://github.com/lcsb-bds/transreg).

stat.ME↗

arXiv2022

Estimation of Predictive Performance in High-Dimensional Data Settings using Learning Curves

In high-dimensional prediction settings, it remains challenging to reliably estimate the test performance. To address this challenge, a novel performance estimation framework is presented. This framework, called Learn2Evaluate, is based on learning curves by fitting a smooth monotone curve depicting test performance as a function of the sample size. Learn2Evaluate has several advantages compared to commonly applied performance estimation methodologies. Firstly, a learning curve offers a graphical overview of a learner. This overview assists in assessing the potential benefit of adding training samples and it provides a more complete comparison between learners than performance estimates at a fixed subsample size. Secondly, a learning curve facilitates in estimating the performance at the total sample size rather than a subsample size. Thirdly, Learn2Evaluate allows the computation of a theoretically justified and useful lower confidence bound. Furthermore, this bound may be tightened by performing a bias correction. The benefits of Learn2Evaluate are illustrated by a simulation study and applications to omics data.

stat.ME↗

arXiv2022

ecpc: An R-package for generic co-data models for high-dimensional prediction

High-dimensional prediction considers data with more variables than samples. Generic research goals are to find the best predictor or to select variables. Results may be improved by exploiting prior information in the form of co-data, providing complementary data not on the samples, but on the variables. We consider adaptive ridge penalised generalised linear and Cox models, in which the variable specific ridge penalties are adapted to the co-data to give a priori more weight to more important variables. The R-package ecpc originally accommodated various and possibly multiple co-data sources, including categorical co-data, i.e. groups of variables, and continuous co-data. Continuous co-data, however, was handled by adaptive discretisation, potentially inefficiently modelling and losing information. Here, we present an extension to the method and software for generic co-data models, particularly for continuous co-data. At the basis lies a classical linear regression model, regressing prior variance weights on the co-data. Co-data variables are then estimated with empirical Bayes moment estimation. After placing the estimation procedure in the classical regression framework, extension to generalised additive and shape constrained co-data models is straightforward. Besides, we show how ridge penalties may be transformed to elastic net penalties with the R-package squeezy. In simulation studies we first compare various co-data models for continuous co-data from the extension to the original method. Secondly, we compare variable selection performance to other variable selection methods. Moreover, we demonstrate use of the package in several examples throughout the paper.

stat.ME↗

arXiv2021

A Bayesian accelerated failure time model for interval censored three-state screening outcomes

Women infected by the Human papilloma virus are at an increased risk to develop cervical intraepithalial neoplasia lesions (CIN). CIN are classified into three grades of increasing severity (CIN-1, CIN-2, and CIN-3) and can eventually develop into cervical cancer. The main purpose of screening is detecting CIN-2 and CIN-3 cases which are usually surgically removed. Screening data from the POBASCAM trial involving 1,454 HPV-positive women is analyzed with two objectives: estimate (a) the transition time from HPV diagnosis to CIN-3; and (b) the transition time from CIN-2 to CIN-3. The screening data have two key characteristics. First, the CIN state is monitored in an interval-censored sequence of screening times. Second, a woman's progression to CIN-3 is only observed, if the woman progresses to, both, CIN-2 and from CIN-2 to CIN-3 in the same screening interval. We propose a Bayesian accelerated failure time model for the two transition times in this three-state model. To deal with the unusual censoring structure of the screening data, we develop a Metropolis-within-Gibbs algorithm with data augmentation from the truncated transition time distributions.

stat.ME↗

arXiv2021

Semi-supervised empirical Bayes group-regularized factor regression

The features in high dimensional biomedical prediction problems are often well described with lower dimensional manifolds. An example is genes that are organised in smaller functional networks. The outcome can then be described with the factor regression model. A benefit of the factor model is that is allows for straightforward inclusion of unlabeled observations in the estimation of the model, i.e., semi-supervised learning. In addition, the high dimensional features in biomedical prediction problems are often well characterised. Examples are genes, for which annotation is available, and metabolites with $p$-values from a previous study available. In this paper, the extra information on the features is included in the prior model for the features. The extra information is weighted and included in the estimation through empirical Bayes, with Variational approximations to speed up the computation. The method is demonstrated in simulations and two applications. One application considers influenza vaccine efficacy prediction based on microarray data. The second application predictions oral cancer metastatsis from RNAseq data.

stat.ME↗

arXiv2021

Fast cross-validation for multi-penalty ridge regression

High-dimensional prediction with multiple data types needs to account for potentially strong differences in predictive signal. Ridge regression is a simple model for high-dimensional data that has challenged the predictive performance of many more complex models and learners, and that allows inclusion of data type specific penalties. The largest challenge for multi-penalty ridge is to optimize these penalties efficiently in a cross-validation (CV) setting, in particular for GLM and Cox ridge regression, which require an additional estimation loop by iterative weighted least squares (IWLS). Our main contribution is a computationally very efficient formula for the multi-penalty, sample-weighted hat-matrix, as used in the IWLS algorithm. As a result, nearly all computations are in low-dimensional space, rendering a speed-up of several orders of magnitude. We developed a flexible framework that facilitates multiple types of response, unpenalized covariates, several performance criteria and repeated CV. Extensions to paired and preferential data types are included and illustrated on several cancer genomics survival prediction problems. Moreover, we present similar computational shortcuts for maximum marginal likelihood and Bayesian probit regression. The corresponding R-package, multiridge, serves as a versatile standalone tool, but also as a fast benchmark for other more complex models and multi-view learners.

stat.ME↗