arXiv subjects

Ernst-Jan Camiel Wit

Publications and source records attributed to Ernst-Jan Camiel Wit.

5 recordsLinked to original sources

arXiv2026

The Regularization Parameter: Sparse Precision Matrix Estimation

Sparse precision matrix estimation provides an interpretable and computationally efficient framework for modeling conditional dependencies in high-dimensional, low-sample-size data. A recurring challenge is appropriately selecting the regularization parameter that controls estimator sparsity and strikes a balance between underfitting and overfitting. We propose a closed-form, matrix-valued regularization parameter derived from the sampling distribution of the first-order optimality conditions of the $\ell_1$-regularized Gaussian maximum-likelihood estimator. By prescribing the probability that each nonzero entry of the estimator satisfies its optimality condition under resampling, we eliminate the need for cross-validation. The resulting regularization parameter is shown to attain asymptotic scaling properties that, under standard conditions, provide consistency and sparsistency of the estimator. On synthetic Gaussian and non-Gaussian datasets, as well as real-world gene microarray and neuroimaging applications, the proposed approach achieves estimation accuracy comparable to cross-validation, delivers superior support recovery, and reduces runtime by several orders of magnitude.

stat.ML↗

arXiv2025

Mixed additive modelling of global alien species co-invasions of plants and insects

Alien species refer to non-native species introduced by humans into an ecosystem, which can cause harm to the environment, economy, or human health. Although there is considerable literature on the subject, the presence of confounding factors has so far prevented a comprehensive picture of the relative importance of various drivers of such invasions. In this manuscript, we aim to develop and apply a general mixed additive relational event model to describe the pattern of global invasions of alien species. The diffusion of alien species can be regarded as a relational event, where the species -- the sender -- reaches a region -- the receiver -- at a specific time in history. We use the First Record Database, which contains all co-invasions by insects and plants between 1880 and 2005. A relational event model (REM) is employed to describe the underlying hazard of each species-region pair. Besides potentially time-varying, exogenous, and endogenous covariates, the mixed additive REM incorporates time-varying and random effects, allowing for taxa-specific baseline rates while accounting for the potential synergistic effect between plants and insects in the invasion process. Our efficient inference procedure relies on case-control sampling, yielding the same likelihood as that of a degenerate logistic regression. We propose fitting the mixed additive REM via a generalised additive model with random effects as 0-dimensional splines. The resulting computational efficiency means that complex models for large dynamic networks can be estimated in seconds on a standard computer. Furthermore, we present a framework for testing the goodness-of-fit of our mixed additive REM for the invasions by vascular plants and insects by means of cumulative martingale-residuals. Implementation is performed through the R package mgcv.

stat.AP↗

arXiv2024

Functional Gaussian Graphical Regression Models For Air Quality Data

Functional data describe a wide range of processes, such as growth curves and spectral absorption. In this study, we analyze air pollution data from the In-service Aircraft for a Global Observing System, focusing on the spatial interactions among chemicals in the atmosphere and their dependence on meteorological conditions. This requires functional regression, where both response and covariates are functional objects evolving over the troposphere. Evaluating both the functional relatedness between the response and covariates and the relatedness of a multivariate response function can be challenging. We propose a solution to these challenges by introducing a functional Gaussian graphical regression model, extending conditional Gaussian graphical models to partially separable functions. To estimate the model, we propose a doubly-penalized estimator. Additionally, we present a novel adaptation of Kullback-Leibler cross-validation tailored for graph estimators which accounts for precision and regression matrices when the population presents one or more sub-groups, named joint Kullback-Leibler cross-validation. Evaluation of model performance is done in terms of Kullback-Leibler divergence and graph recovery power.

stat.ME↗

arXiv2024

Goodness of fit of relational event models

A type of dynamic network involves temporally ordered interactions between actors, where past network configurations may influence future ones. The relational event model can be used to identify the underlying dynamics that drive interactions among system components. Despite the rapid development of this model over the past 15 years, an ongoing area of research revolves around evaluating the goodness of fit of this model, especially when it incorporates time-varying and random effects. Current methodologies often rely on comparing observed and simulated events using specific statistics, but this can be computationally intensive, and requires various assumptions. We propose an additive mixed-effect relational event model estimated via case-control sampling, and introduce a versatile framework for testing the goodness of fit of such models using weighted martingale residuals. Our focus is on a Kolmogorov-Smirnov type test designed to assess if covariates are accurately modeled. Our approach can be easily extended to evaluate whether other features of network dynamics have been appropriately incorporated into the model. We assess the goodness of fit of various relational event models using synthetic data to evaluate the test's power and coverage. Furthermore, we apply the method to a social study involving 57,791 emails sent by 159 employees of a Polish manufacturing company in 2010. The method is implemented in the R package mgcv.

stat.ME↗

arXiv2021

Using Network-based Causal Inference to Detect the Sources of Contagion in the Currency Market

Contagion is an extremely important topic in finance. Contagion is at the core of most major financial crises, in particular the 2008 financial crisis. Although various approaches to quantifying contagion have been proposed, many of them lack a causal interpretation. We will present a new measure for contagion among individual currencies within the Foreign exchange market and show how the paths of contagion work within the Forex using causal inference. This approach will allow us to pinpoint sources of contagion and to find which currencies offer good options for diversification and which are more susceptible to systemic risk, ultimately resulting in feedback on the level of global systemic risk.

q-fin.ST↗