arXiv subjects

Magne Thoresen

Publications and source records attributed to Magne Thoresen.

At least 19 recordsLinked to original sources

arXiv2026

Robust and Scalable Sure Screening of Fixed effects in Ultrahigh-dimensional Linear Mixed Models

In modern applications of linear mixed models, the number of candidate fixed-effects covariates can grow exponentially with the sample size, while dependence induced by random effects and possible data contamination pose substantial challenges for existing variable screening methods. We propose a robust and computationally efficient sure screening procedure for identifying relevant fixed-effects covariates in ultrahigh-dimensional linear mixed models with known random effects. The proposed method leverages a proxy-based transformation to decouple dependence induced by random effects, enabling screening via marginal analysis in a transformed regression model. Robustness is achieved by constructing marginal utilities based on minimum density power divergence, yielding stability under data contamination and model misspecification without sacrificing scalability. The resulting procedure, termed DPD-SISP, is shown to retain all relevant covariates (sure screening property) with exponentially high probability under general conditions, allowing for non-Gaussian errors and nonpolynomial growth of dimensionality. In addition, DPD-SISP exhibits strong robustness properties supported by influence function and breakdown point analyses. The framework is further extended to incorporate prior information through conditional screening, mitigate correlation-induced masking via iterative refinement, and enable robust post-screening estimation of fixed effects. Extensive simulation studies demonstrate competitive performance of DPD-SISP under ideal settings and substantial gains in stability under data contamination. Its practical utility is illustrated through an application to high-dimensional longitudinal data from the ADNI2 study. The proposed framework thus provides a unified, robust, and scalable approach for variable screening in ultrahigh-dimensional linear mixed models.

stat.ME↗

arXiv2026

A Semiparametric Nonlinear Mixed Effects Model with Penalized Splines Using Automatic Differentiation

We present an estimation procedure for nonlinear mixed-effects models in which the population trajectory is represented by penalized splines and adapted to individuals via subject-specific transformation parameters. By exploiting the mixed model representation of penalized splines, the level of smoothness can be estimated jointly with other variance components. The integration over random effects needed to obtain the marginal likelihood is carried out using the Laplace approximation. Exact derivatives for evaluation and maximization of the resulting likelihood are obtained via automatic differentiation implemented through Template Model Builder. In simulation studies, the method produces improved inferential performance and reduced computational burden when compared to the existing procedure. The approach is further illustrated through a case study on infant height growth in the first two years of life.

stat.ME↗

arXiv2025

False Discovery Rate Control via Bayesian Mirror Statistic

Simultaneously performing variable selection and inference in high-dimensional models is an open challenge in statistics and machine learning. The increasing availability of vast amounts of variables requires the adoption of specific statistical procedures to accurately select the most important predictors in a high-dimensional space, while being able to control some form of selection error. In this work we adapt the Mirror Statistic approach to False Discovery Rate (FDR) control into a Bayesian modelling framework. The Mirror Statistic, developed in the classic frequentist statistical framework, is a flexible method to control FDR, which only requires mild model assumptions, but requires two sets of independent regression coefficient estimates, usually obtained after splitting the original dataset. Here we propose to rely on a Bayesian formulation of the model and use the posterior distributions of the coefficients of interest to build the Mirror Statistic and effectively control the FDR without the need to split the data. Moreover, the method is very flexible since it can be used with continuous and discrete outcomes and more complex predictors, such as with mixed models. We keep the approach scalable to high-dimensions by relying on Automatic Differentiation Variational Inference and fully continuous prior choices.

stat.ME↗

arXiv2025

A Computationally Efficient Approach to False Discovery Rate Control and Power Maximisation via Randomisation and Mirror Statistic

Simultaneously performing variable selection and inference in high-dimensional regression models is an open challenge in statistics and machine learning. The increasing availability of vast amounts of variables requires the adoption of specific statistical procedures to accurately select the most important predictors in a high-dimensional space, while controlling the false discovery rate (FDR) associated with the variable selection procedure. In this paper, we propose the joint adoption of the Mirror Statistic approach to FDR control, coupled with outcome randomisation to maximise the statistical power of the variable selection procedure, measured through the true positive rate. Through extensive simulations, we show how our proposed strategy allows us to combine the benefits of the two techniques. The Mirror Statistic is a flexible method to control FDR, which only requires mild model assumptions, but requires two sets of independent regression coefficient estimates, usually obtained after splitting the original dataset. Outcome randomisation is an alternative to data splitting that allows to generate two independent outcomes, which can then be used to estimate the coefficients that go into the construction of the Mirror Statistic. The combination of these two approaches provides increased testing power in a number of scenarios, such as highly correlated covariates and high percentages of active variables. Moreover, it is scalable to very high-dimensional problems, since the algorithm has a low memory footprint and only requires a single run on the full dataset, as opposed to iterative alternatives such as multiple data splitting.

stat.ME↗

arXiv2025

Methods of Selective Inference for Linear Mixed Models: a Review and Empirical Comparison

Selective inference aims at providing valid inference after a data-driven selection of models or hypotheses. It is essential to avoid overconfident results and replicability issues. While significant advances have been made in this area for standard regression models, relatively little attention has been given to linear mixed models (LMMs), which are widely used for analyzing clustered or longitudinal data. This paper reviews the existing selective inference approaches developed for LMMs, focusing on selection of fixed effects, where the random effects structure is given. We present these methods in detail and, through comparative simulations, assess their practical performance and computational feasibility under varying data structures. In addition, we apply them to a real-world biological dataset to examine how method choice can impact inference in practice. Our findings highlight an existing trade-off between computational complexity and statistical power and emphasize the scarcity of methods that perform well as the number of variables increases. In such scenarios, basic sample splitting emerges as the most reliable approach.

stat.ME↗

arXiv2025

Methods for differential network estimation: an empirical comparison

We provide a review and a comparison of methods for differential network estimation in Gaussian graphical models with focus on structure learning. We consider the case of two datasets from distributions associated with two graphical models. In our simulations, we use five different methods to estimate differential networks. We vary graph structure and sparsity to explore their influence on performance in terms of power and false discovery rate. We demonstrate empirically that presence of hubs proves to be a challenge for all the methods, as well as increased density. We suggest local and global properties that are associated with this challenge. Direct estimation with lasso penalized D-trace loss is shown to perform the best across all combinations of network structure and sparsity.

stat.ME↗

arXiv2025

Bad estimation, good prediction: the Lasso in dense regimes

For high-dimensional omics data, sparsity-inducing regularization methods such as the Lasso are widely used and often yield strong predictive performance, even in settings when the assumption of sparsity is likely violated. We demonstrate that under a specific dense model, namely the high-dimensional joint latent variable model, the Lasso produces sparse prediction rules with favorable prediction error bounds, even when the underlying regression coefficient vector is not sparse at all. We further argue that this model better represents many types of omics data than sparse linear regression models. We prove that the prediction bound under this model in fact decreases with increasing number of predictors, and confirm this through simulation examples. These results highlight the need for caution when interpreting sparse prediction rules, as strong prediction accuracy of a sparse prediction rule may not imply underlying biological significance of the individual predictors.

stat.ME↗

arXiv2025

Conditional variable screening for ultra-high dimensional longitudinal data with time interactions

In recent years we have been able to gather large amounts of genomic data at a fast rate, creating situations where the number of variables greatly exceeds the number of observations. In these situations, most models that can handle a moderately high dimension will now become computationally infeasible or unstable. Hence, there is a need for a pre-screening of variables to reduce the dimension efficiently and accurately to a more moderate scale. There has been much work to develop such screening procedures for independent outcomes. However, much less work has been done for high-dimensional longitudinal data in which the observations can no longer be assumed to be independent. In addition, it is of interest to capture possible interactions between the genomic variable and time in many of these longitudinal studies. In this work, we propose a novel conditional screening procedure that ranks variables according to the likelihood value at the maximum likelihood estimates in a marginal linear mixed model, where the genomic variable and its interaction with time are included in the model. This is to our knowledge the first conditional screening approach for clustered data. We prove that this approach enjoys the sure screening property, and assess the finite sample performance of the method through simulations.

stat.ME↗

arXiv2024

Comparison of the LASSO and Integrative LASSO with Penalty Factors (IPF-LASSO) methods for multi-omics data: Variable selection with Type I error control

Variable selection in relation to regression modeling has constituted a methodological problem for more than 60 years. Especially in the context of high-dimensional regression, developing stable and reliable methods, algorithms, and computational tools for variable selection has become an important research topic. Omics data is one source of such high-dimensional data, characterized by diverse genomic layers, and an additional analytical challenge is how to integrate these layers into various types of analyses. While the IPF-LASSO model has previously explored the integration of multiple omics modalities for feature selection and prediction by introducing distinct penalty parameters for each modality, the challenge of incorporating heterogeneous data layers into variable selection with Type I error control remains an open problem. To address this problem, we applied stability selection as a method for variable selection with false positives control in both IPF-LASSO and regular LASSO. The objective of this study was to compare the LASSO algorithm with IPF-LASSO, investigating whether introducing different penalty parameters per omics modality could improve statistical power while controlling false positives. Two high-dimensional data structures were investigated, one with independent data and the other with correlated data. The different models were also illustrated using data from a study on breast cancer treatment, where the IPF-LASSO model was able to select some highly relevant clinical variables.

stat.ME↗

arXiv2024

penalizedclr: an R package for penalized conditional logistic regression for integration of multiple omics layers

The matched case-control design, up until recently mostly pertinent to epidemiological studies, is becoming customary in biomedical applications as well. For instance, in omics studies, it is quite common to compare cancer and healthy tissue from the same patient. Furthermore, researchers today routinely collect data from various and variable sources that they wish to relate to the case-control status. This highlights the need to develop and implement statistical methods that can take these tendencies into account. We present an R package penalizedclr, that provides an implementation of the penalized conditional logistic regression model for analyzing matched case-control studies. It allows for different penalties for different blocks of covariates, and it is therefore particularly useful in the presence of multi-source omics data. Both L1 and L2 penalties are implemented. Additionally, the package implements stability selection for variable selection in the considered regression model. The proposed method fills a gap in the available software for fitting high-dimensional conditional logistic regression model accounting for the matched design and block structure of predictors/features. The output consists of a set of selected variables that are significantly associated with case-control status. These features can then be investigated in terms of functional interpretation or validation in further, more targeted studies.

stat.ME↗

arXiv2023

Exponential Consistency of M-estimators in Generalized Linear Mixed Models

Generalized linear mixed models are powerful tools for analyzing clustered data, where the unknown parameters are classically (and most commonly) estimated by the maximum likelihood and restricted maximum likelihood procedures. However, since the likelihood based procedures are known to be highly sensitive to outliers, M-estimators have become popular as a means to obtain robust estimates under possible data contamination. In this paper, we prove that, for sufficiently smooth general loss functions defining the M-estimators in generalized linear mixed models, the tail probability of the deviation between the estimated and the true regression coefficients have an exponential bound. This implies an exponential rate of consistency of these M-estimators under appropriate assumptions, generalizing the existing exponential consistency results from univariate to multivariate responses. We have illustrated this theoretical result further for the special examples of the maximum likelihood estimator and the robust minimum density power divergence estimator, a popular example of model-based M-estimators, in the settings of linear and logistic mixed models, comparing it with the empirical rate of convergence through simulation studies.

math.ST↗

arXiv2022

Robust Sure Independence Screening for Non-polynomial dimensional Generalized Linear Models

We consider the problem of variable screening in ultra-high dimensional generalized linear models (GLMs) of non-polynomial orders. Since the popular SIS approach is extremely unstable in the presence of contamination and noise, we discuss a new robust screening procedure based on the minimum density power divergence estimator (MDPDE) of the marginal regression coefficients. Our proposed screening procedure performs well under pure and contaminated data scenarios. We provide a theoretical motivation for the use of marginal MDPDEs for variable screening from both population as well as sample aspects; in particular, we prove that the marginal MDPDEs are uniformly consistent leading to the sure screening property of our proposed algorithm. Finally, we propose an appropriate MDPDE based extension for robust conditional screening in GLMs along with the derivation of its sure screening property. Our proposed methods are illustrated through extensive numerical studies along with an interesting real data application.

math.ST↗

arXiv2021

RaJIVE: Robust Angle Based JIVE for Integrating Noisy Multi-Source Data

With increasing availability of high dimensional, multi-source data, the identification of joint and data specific patterns of variability has become a subject of interest in many research areas. Several matrix decomposition methods have been formulated for this purpose, for example JIVE (Joint and Individual Variation Explained), and its angle based variation, aJIVE. Although the effect of data contamination on the estimated joint and individual components has not been considered in the literature, gross errors and outliers in the data can cause instability in such methods, and lead to incorrect estimation of joint and individual variance components. We focus on the aJIVE factorization method and provide a thorough analysis of the effect outliers on the resulting variation decomposition. After showing that such effect is not negligible when all data-sources are contaminated, we propose a robust extension of aJIVE (RaJIVE) that integrates a robust formulation of the singular value decomposition into the aJIVE approach. The proposed RaJIVE is shown to provide correct decompositions even in the presence of outliers and improves the performance of aJIVE. We use extensive simulation studies with different levels of data contamination to compare the two methods. Finally, we describe an application of RaJIVE to a multi-omics breast cancer dataset from The Cancer Genome Atlas. We provide the R package RaJIVE with a ready-to-use implementation of the methods and documentation of code and examples.

stat.ME↗

arXiv2020

Permutation testing in high-dimensional linear models: an empirical investigation

Permutation testing in linear models, where the number of nuisance coefficients is smaller than the sample size, is a well-studied topic. The common approach of such tests is to permute residuals after regressing on the nuisance covariates. Permutation-based tests are valuable in particular because they can be highly robust to violations of the standard linear model, such as non-normality and heteroscedasticity. Moreover, in some cases they can be combined with existing, powerful permutation-based multiple testing methods. Here, we propose permutation tests for models where the number of nuisance coefficients exceeds the sample size. The performance of the novel tests is investigated with simulations. In a wide range of simulation scenarios our proposed permutation methods provided appropriate type I error rate control, unlike some competing tests, while having good power.

stat.ME↗

arXiv2020

On optimal two-stage testing of multiple mediators

Mediation analysis in high-dimensional settings often involves identifying potential mediators among a large number of measured variables. For this purpose, a two-step familywise error rate procedure called ScreenMin has been recently proposed (Djordjilović et al. 2019). In ScreenMin, variables are first screened and only those that pass the screening are tested. The proposed threshold for selection has been shown to guarantee asymptotic familywise error rate. In this work, we investigate the impact of the selection threshold on the finite sample familywise error rate. We derive a power maximizing selection threshold and show that it is well approximated by an adaptive threshold of Wang et al. (2016). We illustrate the investigated procedures on a case-control study examining the effect of fish intake on the risk of colorectal adenoma.

stat.ME↗

arXiv2020

When and why are principal component scores a good tool for visualizing high-dimensional data?

Principal component analysis (PCA) is a popular dimension reduction technique often used to visualize high-dimensional data structures. In genomics, this can involve millions of variables, but only tens to hundreds of observations. Theoretically, such extreme high-dimensionality will cause biased or inconsistent eigenvector estimates, but in practice the principal component scores are used for visualization with great success. In this paper, we explore when and why the classical principal component scores can be used to visualize structures in high-dimensional data, even when there are few observations compared to the number of variables. Our argument is two-fold: First, we argue that eigenvectors related to pervasive signals will have eigenvalues scaling linearly with the number of variables. Second, we prove that for linearly increasing eigenvalues, the sample component scores will be scaled and rotated versions of the population scores, asymptotically. Thus the visual information of the sample scores will be unchanged, even though the sample eigenvectors are biased. In the case of pervasive signals, the principal component scores can be used to visualize the population structures, even in extreme high-dimensional situations.

math.ST↗

arXiv2020

A robust variable screening procedure for ultra-high dimensional data

Variable selection in ultra-high dimensional regression problems has become an important issue. In such situations, penalized regression models may face computational problems and some pre screening of the variables may be necessary. A number of procedures for such pre-screening has been developed; among them the sure independence screening (SIS) enjoys some popularity. However, SIS is vulnerable to outliers in the data, and in particular in small samples this may lead to faulty inference. In this paper, we develop a new robust screening procedure. We build on the density power divergence (DPD) estimation approach and introduce DPD-SIS and its extension iterative DPD-SIS. We illustrate the behavior of the methods through extensive simulation studies and show that they are superior to both the original SIS and other robust methods when there are outliers in the data. We demonstrate the claimed robustness through use of influence functions, and we discuss appropriate choice of the tuning parameter $α$. Finally, we illustrate its use on a small dataset from a study on regulation of lipid metabolism.

stat.ME↗

arXiv2020

Consistent Fixed-Effects Selection in Ultra-high dimensional Linear Mixed Models with Error-Covariate Endogeneity

Recently, applied sciences, including longitudinal and clustered studies in biomedicine require the analysis of ultra-high dimensional linear mixed effects models where we need to select important fixed effect variables from a vast pool of available candidates. However, all existing literature assume that all the available covariates and random effect components are independent of the model error which is often violated (endogeneity) in practice. In this paper, we first investigate this important issue in ultra-high dimensional linear mixed effects models with particular focus on the fixed effects selection. We study the effects of different types of endogeneity on existing regularization methods and prove their inconsistencies. Then, we propose a new profiled focused generalized method of moments (PFGMM) approach to consistently select fixed effects under 'error-covariate' endogeneity, i.e., in the presence of correlation between the model error and covariates. Our proposal is proved to be oracle consistent with probability tending to one and works well under most other type of endogeneity too. Additionally, we also propose and illustrate a few consistent parameter estimators, including those of the variance components, along with variable selection through PFGMM. Empirical simulations and an interesting real data example further support the claimed utility of our proposal.

stat.ME↗