Searcharxiv⌕ Search

arXiv subjects

Myeonghun Yu

Publications and source records attributed to Myeonghun Yu.

3 recordsLinked to original sources

Large-scale linear hypothesis testing for high-dimensional online M-estimation

With the growing prevalence of streaming data, renewable estimation has become an important tool for updating statistical analyses without retaining historical raw observations. We study estimation and linear hypothesis testing in high-dimensional online M-estimation. Existing renewable procedures based on quadratic approximations can incur first-order approximation errors because the gradient of the unpenalized loss generally does not vanish at a preceding penalized estimator, posing a particular challenge for valid inference. To address this issue, we propose a corrected renewable loss that preserves historical first-order information and combine it with partial folded concave regularization and an iterative local linear approximation. We establish finite sample $\ell_1$- and $\ell_2$-error bounds, together with contraction bounds and strong oracle properties for both the constrained and unconstrained iterative estimators. Building on these results, we derive oracle Bahadur representations and establish asymptotically valid Wald and score tests for general linear hypotheses, allowing both the testing-oracle dimension and the number of restrictions to diverge. We further characterize their asymptotic power under local alternatives. The proposed procedure requires only recursively updated summaries while recovering the first-order inferential behavior of its pooled-data oracle counterpart. Simulation studies and real-data applications demonstrate the finite-sample performance of the proposed estimation and testing procedures.

stat.ME↗

Deep neural expected shortfall regression with tail-robustness

Expected shortfall (ES), also known as conditional value-at-risk, is a widely recognized risk measure that complements value-at-risk by capturing tail-related risks more effectively. Compared with quantile regression, which has been extensively developed and applied across disciplines, ES regression remains in its early stage, partly because the traditional empirical risk minimization framework is not directly applicable. In this paper, we develop a nonparametric framework for expected shortfall regression based on a two-step approach that treats the conditional quantile function as a nuisance parameter. Leveraging the representational power of deep neural networks, we construct a two-step ES estimator using feedforward ReLU networks, which can alleviate the curse of dimensionality when the underlying functions possess hierarchical composition structures. However, ES estimation is inherently sensitive to heavy-tailed response or error distributions. To address this challenge, we integrate a properly tuned Huber loss into the neural network training, yielding a robust deep ES estimator that is provably resistant to heavy-tailedness in a non-asymptotic sense and first-order insensitive to quantile estimation errors in the first stage. Comprehensive simulation studies and an empirical analysis of the effect of El Niño on extreme precipitation illustrate the accuracy and robustness of the proposed method.

stat.ME↗

Fortified Proximal Causal Inference with Many Invalid Proxies

Causal inference from observational data often relies on the assumption of no unmeasured confounding, an assumption frequently violated in practice due to unobserved or poorly measured covariates. Proximal causal inference (PCI) offers a promising framework for addressing unmeasured confounding using a pair of outcome and treatment confounding proxies. However, existing PCI methods typically assume all specified proxies are valid, which may be unrealistic and is untestable without extra assumptions. In this paper, we develop a semiparametric approach for a many-proxy PCI setting that accommodates potentially invalid treatment confounding proxies. We introduce a new class of fortified confounding bridge functions and establish nonparametric identification of the population average treatment effect (ATE) under the assumption that at least $γ$ out of $K$ candidate treatment confounding proxies are valid, for any $γ\leq K$ set by the analyst without requiring knowledge of which proxies are valid. We establish a local semiparametric efficiency bound and develop a class of multiply robust, locally efficient estimators for the ATE. These estimators are thus simultaneously robust to invalid treatment confounding proxies and model misspecification of nuisance parameters. The proposed methods are evaluated through simulation and applied to assess the effect of right heart catheterization in critically ill patients.

stat.ME↗