arXiv subjects

Thomas Nagler

Publications and source records attributed to Thomas Nagler.

At least 19 recordsLinked to original sources

arXiv2026

Asymptotics for estimating a diverging number of parameters -- with and without sparsity

We develop a general asymptotic theory for estimating equations whose dimension diverges with the sample size. For both unpenalized and sparse penalized problems, we establish population-level conditions for existence, consistency, uniqueness, and asymptotic normality; in the penalized setting, we also establish selection consistency under a generalized version of the mutual incoherence condition. Our results cover stepwise procedures with a diverging number of steps, independent but non-identically distributed observations, and dependent data. We allow penalties that are simultaneously nonconvex, non-coordinate-separable, and group-structured and may involve heterogeneous tuning parameters. Our population-level conditions imply a weak form of restricted strong convexity, and we provide an explicit example where the commonly used stronger form fails. The results are illustrated by several applications including Group SCAD-penalized estimation in generalized linear models, distributed inference under network dependence, and penalized stepwise estimation in causal inference.

math.ST↗

arXiv2026

Online Bootstrap Inference for the Trend of Nonstationary Time Series

This article proposes an online bootstrap scheme for nonparametric level estimation in nonstationary time series. Our approach applies to a broad class of level estimators expressible as weighted sample averages over time windows, including exponential smoothing methods and moving averages. The bootstrap procedure is motivated by asymptotic arguments and provides well-calibrated uniform-in-time coverage, enabling scalable uncertainty quantification in streaming or large-scale time-series settings. This makes the method suitable for tasks such as adaptive anomaly detection, online monitoring, or streaming A/B testing. Simulation studies demonstrate good finite-sample performance of our method across a range of nonstationary scenarios. In summary, this offers a practical resampling framework that complements online trend estimation with reliable statistical inference.

stat.ME↗

arXiv2026

Mortality Forecasting under Climate Risk: A Stochastic Approach with Distributed Lag Non-Linear Models

Assessing climate-driven mortality risk has become an emerging area of research in recent decades. In this paper, we propose a novel approach to explicitly incorporate climate-driven effects into both single- and multi-population stochastic mortality models. The new model consists of two components: a stochastic mortality model, and a distributed lag non-linear model (DLNM). The stochastic component captures the non-climate long-term trend, volatility, and seasonal patterns in mortality rates. The DLNM component captures non-linear and lagged effects of climate variables on mortality, as well as the impact of heat waves and cold waves across different age groups. For model calibration, we propose a novel backfitting algorithm that allows us to disentangle the climate-driven mortality risk from the non-climate-driven stochastic mortality risk. We illustrate the effectiveness and improved short-term (1--18 month) forecasting performance of our model against four alternative models, using data from three European regions: Athens, Lisbon, and Rome. Furthermore, as an application of the proposed modeling framework, we utilize future UTCI data generated from climate models to provide total mortality forecasts into 2045 across these regions under two Representative Concentration Pathway (RCP) scenarios, taking both stochastic mortality improvement trend and climate risk into account. The projections show a noticeable decrease in winter mortality alongside a rise in summer mortality, driven by a general increase in UTCI over time. Although we expect slightly lower overall mortality in the short term under RCP8.5 compared to RCP2.6, a long-term increase in total mortality is anticipated under the RCP8.5 scenario.

stat.AP↗

arXiv2026

Uniform central limit theorems for non-stationary processes via relative weak convergence

Statistical inference for non-stationary data is hindered by the failure of classical central limit theorems (CLTs), not least because there is no fixed Gaussian limit to converge to. To resolve this, we introduce relative weak convergence, an extension of weak convergence that compares a statistic or process to a sequence of evolving processes. Relative weak convergence retains the essential consequences of classical weak convergence and coincides with it under stationarity. Crucially, it applies in general non-stationary settings where classical weak convergence fails. We establish concrete relative CLTs for random vectors and empirical processes, along with sequential, weighted, and bootstrap variants that parallel the state-of-the-art in stationary settings. Our framework and results offer simple, plug-in replacements for classical CLTs whenever stationarity is untenable, as illustrated by applications in nonparametric trend estimation and hypothesis testing.

math.ST↗

arXiv2026

Properties of stepwise parameter estimation in high-dimensional vine copulas

The increasing use of vine copulas in high-dimensional settings, where the number of parameters is often of the same order as the sample size, calls for asymptotic theory beyond the traditional fixed-$p$, large-$n$ framework. We establish consistency and asymptotic normality of the stepwise maximum likelihood estimator for vine copulas when the number of parameters diverges as $n \to \infty$. Our theoretical results cover both parametric and nonparametric estimation of the marginal distributions, as well as truncated vines, and are also applicable to general estimation problems, particularly other sequential procedures. Numerical experiments suggest that the derived assumptions are satisfied if the pair copulas in higher trees converge to independence copulas sufficiently fast. A simulation study substantiates these findings and identifies settings in which estimation becomes challenging. In particular, the vine structure strongly affects estimation accuracy, with D-vines being more difficult to estimate than C-vines, and estimates in Gumbel vines exhibit substantially larger biases than those in Gaussian vines.

math.ST↗

arXiv2026

Throwing Vines at the Wall: Structure Learning via Random Search

Vine copulas offer flexible multivariate dependence modeling and have become widely used in machine learning. Yet, structure learning remains a key challenge. Early heuristics, such as Dissmann's greedy algorithm, are still considered the gold standard but are often suboptimal. We propose random search algorithms and a statistical framework based on model confidence sets, to improve structure selection, provide theoretical guarantees on selection probabilities and excess risk, as well as serve as a foundation for ensembling. Empirical results on real-world data sets show that our methods consistently outperform state-of-the-art approaches.

stat.ME↗

arXiv2026

Fast Rates for Nonstationary Weighted Risk Minimization

Weighted empirical risk minimization is a common approach to prediction under distribution drift. This article studies its out-of-sample prediction error under nonstationarity. We provide a general decomposition of the excess risk into a learning term and an error term associated with distribution drift, and prove oracle inequalities for the learning error under mixing conditions. The learning bound holds uniformly over arbitrary weight classes and accounts for the effective sample size induced by the weight vector, the complexity of the weight and hypothesis classes, and potential data dependence. We illustrate the applicability and sharpness of our results in (auto-) regression problems with linear models, basis approximations, and neural networks, recovering minimax-optimal rates (up to logarithmic factors) when specialized to unweighted and stationary settings.

stat.ML↗

arXiv2026

Uncertainty Quantification for Prior-Data Fitted Networks using Martingale Posteriors

Prior-data fitted networks (PFNs) have emerged as promising foundation models for prediction from tabular datasets, achieving state-of-the-art performance on small to moderate data sizes without tuning. While PFNs are motivated by Bayesian ideas, they do not provide any uncertainty quantification for predictive means, quantiles, or similar quantities. We propose a principled, efficient, and tuning-free sampling procedure to construct Bayesian posteriors for such estimates based on martingale posteriors, and prove its convergence. Several simulated and real-world data examples showcase the efficiency and calibration of our method in inference applications.

stat.ME↗

arXiv2026

Optimal neural network approximation of smooth compositional functions on sets with low intrinsic dimension

We study approximation and statistical learning properties of deep ReLU networks under structural assumptions that mitigate the curse of dimensionality. We prove minimax-optimal uniform approximation rates for $s$-Hölder smooth functions defined on sets with low Minkowski dimension using fully connected networks with flexible width and depth, improving existing results by logarithmic factors even in classical full-dimensional settings. A key technical ingredient is a new memorization result for deep ReLU networks that enables efficient point fitting with dense architectures. We further introduce a class of compositional models in which each component function is smooth and acts on a domain of low intrinsic dimension. This framework unifies two common assumptions in the statistical learning literature, structural constraints on the target function and low dimensionality of the covariates, within a single model. We show that deep networks can approximate such functions at rates determined by the most difficult function in the composition. As an application, we derive improved convergence rates for empirical risk minimization in nonparametric regression that adapt to smoothness, compositional structure, and intrinsic dimensionality.

math.ST↗

arXiv2025

Adjustment for Confounding using Pre-Trained Representations

There is growing interest in extending average treatment effect (ATE) estimation to incorporate non-tabular data, such as images and text, which may act as sources of confounding. Neglecting these effects risks biased results and flawed scientific conclusions. However, incorporating non-tabular data necessitates sophisticated feature extractors, often in combination with ideas of transfer learning. In this work, we investigate how latent features from pre-trained neural networks can be leveraged to adjust for sources of confounding. We formalize conditions under which these latent features enable valid adjustment and statistical inference in ATE estimation, demonstrating results along the example of double machine learning. We discuss critical challenges inherent to latent feature learning and downstream parameter estimation arising from the high dimensionality and non-identifiability of representations. Common structural assumptions for obtaining fast convergence rates with additive or sparse linear models are shown to be unrealistic for latent features. We argue, however, that neural networks are largely insensitive to these issues. In particular, we show that neural networks can achieve fast convergence rates by adapting to intrinsic notions of sparsity and dimension of the learning problem.

stat.ML↗

arXiv2025

Vine Copulas as Differentiable Computational Graphs

Vine copulas are sophisticated models for multivariate distributions and are increasingly used in machine learning. To facilitate their integration into modern ML pipelines, we introduce the vine computational graph, a DAG that abstracts the multilevel vine structure and associated computations. On this foundation, we devise new algorithms for conditional sampling, efficient sampling-order scheduling, and constructing vine structures for customized conditioning variables. We implement these ideas in torchvinecopulib, a GPU-accelerated Python library built upon PyTorch, delivering improved scalability for fitting, sampling, and density evaluation. Our experiments illustrate how gradient flowing through the vine can improve Vine Copula Autoencoders and that incorporating vines for uncertainty quantification in deep learning can outperform MC-dropout, deep ensembles, and Bayesian Neural Networks in sharpness, calibration, and runtime. By recasting vine copula models as computational graphs, our work connects classical dependence modeling with modern deep-learning toolchains and facilitates the integration of state-of-the-art copula methods in modern machine learning pipelines.

cs.LG↗

arXiv2025

Hybrid Bernstein Normalizing Flows for Flexible Multivariate Density Regression with Interpretable Marginals

Density regression models allow a comprehensive understanding of data by modeling the complete conditional probability distribution. While flexible estimation approaches such as normalizing flows (NF) work particularly well in multiple dimensions, interpreting the input-output relationship of such models is often difficult, due to the black-box character of deep learning models. In contrast, existing statistical methods for multivariate outcomes such as multivariate conditional transformation models (MCTM) are restricted in flexibility and are often not expressive enough to represent complex multivariate probability distributions. In this paper, we combine MCTM with state-of-the-art and autoregressive NF to leverage the transparency of MCTM for modeling interpretable feature effects on the marginal distributions in the first step and the flexibility of neural-network-based NF techniques to account for complex and non-linear relationships in the joint data distribution. We demonstrate our method's versatility in various numerical experiments and compare it with MCTM and other NF models on both simulated and real-world data.

stat.ML↗

arXiv2025

Towards more realistic climate model outputs: A multivariate bias correction based on zero-inflated vine copulas

Climate model large ensembles are an essential research tool for analysing and quantifying natural climate variability and providing robust information for rare extreme events. The models simulated representations of reality are susceptible to bias due to incomplete understanding of physical processes. This paper aims to correct the bias of five climate variables from the CRCM5 Large Ensemble over Central Europe at a 3-hourly temporal resolution. At this high temporal resolution, two variables, precipitation and radiation, exhibit a high share of zero inflation. We propose a novel bias-correction method, VBC (Vine copula bias correction), that models and transfers multivariate dependence structures for zero-inflated margins in the data from its error-prone model domain to a reference domain. VBC estimates the model and reference distribution using vine copulas and corrects the model distribution via (inverse) Rosenblatt transformation. To deal with the variables' zero-inflated nature, we develop a new vine density decomposition that accommodates such variables and employs an adequately randomized version of the Rosenblatt transform. This novel approach allows for more accurate modelling of multivariate zero-inflated climate data. Compared with state-of-the-art correction methods, VBC is generally the best-performing correction and the most accurate method for correcting zero-inflated events.

stat.AP↗

arXiv2025

On dimension reduction in conditional dependence models

Inference of the conditional dependence structure is challenging when many covariates are present. In numerous applications, only a low-dimensional projection of the covariates influences the conditional distribution. The smallest subspace that captures this effect is called the central subspace in the literature. We show that inference of the central subspace of a vector random variable $\mathbf Y$ conditioned on a vector of covariates $\mathbf X$ can be separated into inference of the marginal central subspaces of the components of $\mathbf Y$ conditioned on $\mathbf X$ and on the copula central subspace, that we define in this paper. Further discussion addresses sufficient dimension reduction subspaces for conditional association measures. An adaptive nonparametric method is introduced for estimating the central dependence subspaces, achieving parametric convergence rates under mild conditions. Simulation studies illustrate the practical performance of the proposed approach.

stat.ME↗

arXiv2025

Simplified vine copula models: state of science and affairs

Vine copula models have become highly popular practical tools for modeling multivariate dependencies. To maintain tractability, a commonly employed simplifying assumption is that conditional copulas remain unchanged by the conditioning variables. This assumption has sparked a somewhat polarizing debate within the copula community. The fact that much of this dispute occurs outside the public record has placed the field in an unfortunate position, impeding scientific progress. In this article, I will review what we know about the flexibility and limitations of simplified vine copula models, explore the broader implications, and offer my own, hopefully reconciling, perspective on the issue.

stat.ME↗

arXiv2025

Constructing Confidence Intervals for 'the' Generalization Error -- a Comprehensive Benchmark Study

When assessing the quality of prediction models in machine learning, confidence intervals (CIs) for the generalization error, which measures predictive performance, are a crucial tool. Luckily, there exist many methods for computing such CIs and new promising approaches are continuously being proposed. Typically, these methods combine various resampling procedures, most popular among them cross-validation and bootstrapping, with different variance estimation techniques. Unfortunately, however, there is currently no consensus on when any of these combinations may be most reliably employed and how they generally compare. In this work, we conduct a large-scale study comparing CIs for the generalization error, the first one of such size, where we empirically evaluate 13 different CI methods on a total of 19 tabular regression and classification problems, using seven different inducers and a total of eight loss functions. We give an overview of the methodological foundations and inherent challenges of constructing CIs for the generalization error and provide a concise review of all 13 methods in a unified framework. Finally, the CI methods are evaluated in terms of their relative coverage frequency, width, and runtime. Based on these findings, we can identify a subset of methods that we would recommend. We also publish the datasets as a benchmarking suite on OpenML and our code on GitHub to serve as a basis for further studies.

stat.ML↗

arXiv2024

Decomposing Global Feature Effects Based on Feature Interactions

Global feature effect methods, such as partial dependence plots, provide an intelligible visualization of the expected marginal feature effect. However, such global feature effect methods can be misleading, as they do not represent local feature effects of single observations well when feature interactions are present. We formally introduce generalized additive decomposition of global effects (GADGET), which is a new framework based on recursive partitioning to find interpretable regions in the feature space such that the interaction-related heterogeneity of local feature effects is minimized. We provide a mathematical foundation of the framework and show that it is applicable to the most popular methods to visualize marginal feature effects, namely partial dependence, accumulated local effects, and Shapley additive explanations (SHAP) dependence. Furthermore, we introduce and validate a new permutation-based interaction detection procedure that is applicable to any feature effect method that fits into our proposed framework. We empirically evaluate the theoretical characteristics of the proposed methods based on various feature effect methods in different experimental settings. Moreover, we apply our introduced methodology to three real-world examples to showcase their usefulness.

stat.ML↗

arXiv2024

Reshuffling Resampling Splits Can Improve Generalization of Hyperparameter Optimization

Hyperparameter optimization is crucial for obtaining peak performance of machine learning models. The standard protocol evaluates various hyperparameter configurations using a resampling estimate of the generalization error to guide optimization and select a final hyperparameter configuration. Without much evidence, paired resampling splits, i.e., either a fixed train-validation split or a fixed cross-validation scheme, are often recommended. We show that, surprisingly, reshuffling the splits for every configuration often improves the final model's generalization performance on unseen data. Our theoretical analysis explains how reshuffling affects the asymptotic behavior of the validation loss surface and provides a bound on the expected regret in the limiting regime. This bound connects the potential benefits of reshuffling to the signal and noise characteristics of the underlying optimization problem. We confirm our theoretical results in a controlled simulation study and demonstrate the practical usefulness of reshuffling in a large-scale, realistic hyperparameter optimization experiment. While reshuffling leads to test performances that are competitive with using fixed splits, it drastically improves results for a single train-validation holdout protocol and can often make holdout become competitive with standard CV while being computationally cheaper.

stat.ML↗