Searcharxiv⌕ Search

arXiv subjects

Chenxin Lyu

Publications and source records attributed to Chenxin Lyu.

3 recordsLinked to original sources

On the false discovery rate of one-sided p-values under Gaussian dependence

We study the false discovery rate (FDR) of the Benjamini-Hochberg procedure for $K$ hypotheses at nominal level $q$ under Gaussian dependence of one-sided p-values. We ask whether the global null is least favorable after optimizing over correlation matrices. We show that the global null attains the worst-case FDR for $K\le3$ at every level. In dimension eight, an explicit mixed-null construction has FDR $21q/20$ for $0<q<1/2$. It exceeds the global-null maximum for all sufficiently small $q$, by a uniform extension of the known fixed-matrix small-level Simes bound. For $0<q<1/2$, we identify eight as the first inflation dimension in the monotone rank-one class and prove that $21q/20$ is its exact eight-dimensional maximum. In the same level range, the sharp multiplicative bound over all dimensions in this class is twice the omega constant $w = W(1)$, where $W$ is the Lambert $W$ function. This bound is approached as the dimension increases. We also give scale conditions for stability around the independence benchmark $m_0q/K$, where $m_0$ is the number of true nulls, and obtain a dimension-free upper bound of $q+q^2/2+O(q^3)$ at small levels under the global null with nonpositive correlations.

math.ST↗

Adaptive Window Selection for Financial Risk Forecasting

Risk forecasts in financial regulation and internal management are calculated through historical data. The unknown structural changes of financial data pose a substantial challenge in selecting an appropriate look-back window for risk modeling and forecasting. We develop a data-driven online learning method, called the bootstrap-based adaptive window selection (BAWS), that adaptively determines the window size in a sequential manner. A central component of BAWS is to compare the realized scores against a data-dependent threshold based on the bootstrap method. We provide an asymptotic justification for the bootstrap threshold, covering non-smooth scores such as the VaR check loss and the joint VaR--ES score, with an extension to stationary weakly dependent data via the moving block bootstrap. A single-break analysis further shows that BAWS rejects overlong windows crossing sufficiently large breaks. The proposed method is applicable to the forecasting of risk measures that are elicitable individually or jointly, such as the Value-at-Risk (VaR) and the pair of VaR and the corresponding Expected Shortfall. Through simulation studies and an empirical analysis, we demonstrate that BAWS often improves upon the standard rolling window approach and the recently developed method of stability-based adaptive window selection, especially when there are structural changes in the data-generating process.

q-fin.RM↗

IntOPE: Off-Policy Evaluation in the Presence of Interference

Off-Policy Evaluation (OPE) is employed to assess the potential impact of a hypothetical policy using logged contextual bandit feedback, which is crucial in areas such as personalized medicine and recommender systems, where online interactions are associated with significant risks and costs. Traditionally, OPE methods rely on the Stable Unit Treatment Value Assumption (SUTVA), which assumes that the reward for any given individual is unaffected by the actions of others. However, this assumption often fails in real-world scenarios due to the presence of interference, where an individual's reward is affected not just by their own actions but also by the actions of their peers. This realization reveals significant limitations of existing OPE methods in real-world applications. To address this limitation, we propose IntIPW, an IPW-style estimator that extends the Inverse Probability Weighting (IPW) framework by integrating marginalized importance weights to account for both individual actions and the influence of adjacent entities. Extensive experiments are conducted on both synthetic and real-world data to demonstrate the effectiveness of the proposed IntIPW method.

cs.LG↗