arXiv subjects

Zhonglei Wang

Publications and source records attributed to Zhonglei Wang.

15 recordsLinked to original sources

arXiv2025

DPSformer: A long-tail-aware model for improving heavy rainfall prediction

Accurate and timely forecasting of heavy rainfall remains a critical challenge for modern society. Precipitation exhibits a highly imbalanced distribution: most observations record no or light rain, while heavy rainfall events are rare. Such an imbalanced distribution obstructs deep learning models from effectively predicting heavy rainfall events. To address this challenge, we treat rainfall forecasting explicitly as a long-tailed learning problem, identifying the insufficient representation of heavy rainfall events as the primary barrier to forecasting accuracy. Therefore, we introduce DPSformer, a long-tail-aware model that enriches representation of heavy rainfall events through a high-resolution branch. For heavy rainfall events $ \geq $ 50 mm/6 h, DPSformer lifts the Critical Success Index (CSI) of a baseline Numerical Weather Prediction (NWP) model from 0.012 to 0.067. For the top 1% coverage of heavy rainfall events, its Fraction Skill Score (FSS) exceeds 0.45, surpassing existing methods. Our work establishes an effective long-tailed paradigm for heavy rainfall prediction, offering a practical tool to enhance early warning systems and mitigate the societal impacts of extreme weather events.

cs.LG↗

arXiv2025

Towards Long-Range ENSO Prediction with an Explainable Deep Learning Model

El Niño-Southern Oscillation (ENSO) is a prominent mode of interannual climate variability with far-reaching global impacts. Its evolution is governed by intricate air-sea interactions, posing significant challenges for long-term prediction. In this study, we introduce CTEFNet, a multivariate deep learning model that synergizes convolutional neural networks and transformers to enhance ENSO forecasting. By integrating multiple oceanic and atmospheric predictors, CTEFNet extends the effective forecast lead time to 20 months while mitigating the impact of the spring predictability barrier, outperforming both dynamical models and state-of-the-art deep learning approaches. Furthermore, CTEFNet offers physically meaningful and statistically significant insights through gradient-based sensitivity analysis, revealing the key precursor signals that govern ENSO dynamics, which align with well-established theories and reveal new insights about inter-basin interactions among the Pacific, Atlantic, and Indian Oceans. The CTEFNet's superior predictive skill and interpretable sensitivity assessments underscore its potential for advancing climate prediction. Our findings highlight the importance of multivariate coupling in ENSO evolution and demonstrate the promise of deep learning in capturing complex climate dynamics with enhanced interpretability.

physics.geo-ph↗

arXiv2025

YingLong-weather: AI-Based Limited Area Models for Forecasting of Non-precipitation Surface Meteorological Variables

Recently, artificial intelligence-based (AI-based) models for forecasting of global weather have been rapidly developed. Most of the global models are trained on reanalysis datasets with a spatial resolution of 0.25°*0.25°. However, research on AI-based high spatial resolution limited area weather forecasting models remains limited. In this study, YingLong, an AI-based limited area weather forecasting model with a spatial resolution of 3 km * 3 km is developed. YingLong employs a parallel structure of global and local blocks to capture multiscale meteorological features and operates much faster than the dynamical limited area model WRF-ARW. In two selected limited areas (one relatively flat and the other featuring significant mountain ranges), YingLong (with lateral boundary condition imposed by the global AI-based model Pangu-weather) demonstrates superior skill in forecasting surface wind speed compared to WRF-ARW. Additionally, it shows comparable skill in forecasting surface temperature and pressure. The accuracy of surface temperature and humidity forecasts can be further improved by applying better boundary conditions. YingLong also addresses issues related to the lateral boundary conditions of AI-based limited area models, such as selecting the width of the lateral boundary region and combining finer and coarser resolution predictions in this region. Therefore, YingLong has a great potential to generate cost-effective multiyear high-resolution synthetic wind speed that maintain meteorological realism both spatially and temporally, aiding in the planning and operations for wind power generation companies.

physics.ao-ph↗

arXiv2024

Probability Weighted Clustered Coefficients Regression Models in Complex Survey Sampling

Regression analysis is commonly conducted in survey sampling. However, existing methods fail when the relationships vary across different areas or domains. In this paper, we propose a unified framework to study the group-wise covariate effect under complex survey sampling based on pairwise penalties, and the associated objective function is solved by the alternating direction method of multipliers. Theoretical properties of the proposed method are investigated under some generality conditions. Numerical experiments demonstrate the superiority of the proposed method in terms of identifying groups and estimation efficiency for both linear regression models and logistic regression models.

stat.ME↗

arXiv2024

Mixed Matrix Completion in Complex Survey Sampling under Heterogeneous Missingness

Modern surveys with large sample sizes and growing mixed-type questionnaires require robust and scalable analysis methods. In this work, we consider recovering a mixed dataframe matrix, obtained by complex survey sampling, with entries following different canonical exponential distributions and subject to heterogeneous missingness. To tackle this challenging task, we propose a two-stage procedure: in the first stage, we model the entry-wise missing mechanism by logistic regression, and in the second stage, we complete the target parameter matrix by maximizing a weighted log-likelihood with a low-rank constraint. We propose a fast and scalable estimation algorithm that achieves sublinear convergence, and the upper bound for the estimation error of the proposed method is rigorously derived. Experimental results support our theoretical claims, and the proposed estimator shows its merits compared to other existing methods. The proposed method is applied to analyze the National Health and Nutrition Examination Survey data.

stat.ME↗

arXiv2023

Multiple bias-calibration for adjusting selection bias of non-probability samples using data integration

Valid statistical inference is challenging when the sample is subject to unknown selection bias. Data integration can be used to correct for selection bias when we have a parallel probability sample from the same population with some common measurements. How to model and estimate the selection probability or the propensity score (PS) of a non-probability sample using an independent probability sample is the challenging part of the data integration. We approach this difficult problem by employing multiple candidate models for PS combined with empirical likelihood. By incorporating multiple propensity score models into the internal bias calibration constraint in the empirical likelihood setup, the selection bias can be eliminated so long as the multiple candidate models contain a true PS model. The bias calibration constraint under the multiple PS models is called multiple bias calibration. Multiple PS models can include both missing-at-random and missing-not-at-random models. Asymptotic properties are discussed, and some limited simulation studies are presented to compare the proposed method with some existing competitors. Plasmode simulation studies using the Culture \& Community in a Time of Crisis dataset demonstrate the practical usage and advantages of the proposed method.

stat.ME↗

arXiv2023

Transductive Matrix Completion with Calibration for Multi-Task Learning

Multi-task learning has attracted much attention due to growing multi-purpose research with multiple related data sources. Moreover, transduction with matrix completion is a useful method in multi-label learning. In this paper, we propose a transductive matrix completion algorithm that incorporates a calibration constraint for the features under the multi-task learning framework. The proposed algorithm recovers the incomplete feature matrix and target matrix simultaneously. Fortunately, the calibration information improves the completion results. In particular, we provide a statistical guarantee for the proposed algorithm, and the theoretical improvement induced by calibration information is also studied. Moreover, the proposed algorithm enjoys a sub-linear convergence rate. Several synthetic data experiments are conducted, which show the proposed algorithm out-performs other existing methods, especially when the target matrix is associated with the feature matrix in a nonlinear way.

stat.ML↗

arXiv2022

Survey data integration for regression analysis using model calibration

We consider regression analysis in the context of data integration. To combine partial information from external sources, we employ the idea of model calibration which introduces a "working" reduced model based on the observed covariates. The working reduced model is not necessarily correctly specified but can be a useful device to incorporate the partial information from the external data. The actual implementation is based on a novel application of the information projection and model calibration weighting. The proposed method is particularly attractive for combining information from several sources with different missing patterns. The proposed method is applied to a real data example combining survey data from Korean National Health and Nutrition Examination Survey and big data from National Health Insurance Sharing Service in Korea.

stat.ME↗

arXiv2022

Nonparametric augmented probability weighting with sparsity

Nonresponse frequently arises in practice, and simply ignoring it may lead to erroneous inference. Besides, the number of collected covariates may increase as the sample size in modern statistics, so parametric imputation or propensity score weighting usually leads to inefficiency without consideration of sparsity. In this paper, we propose a nonparametric imputation method with sparse learning by employing an efficient kernel-based learning gradient algorithm to identify truly informative covariates. Moreover, an augmented probability weighting framework is adopted to improve the estimation efficiency of the nonparametric imputation method and establish the limiting distribution of the corresponding estimator under regularity assumptions. The performance of the proposed method is also supported by several simulated examples and one real-life analysis.

stat.ME↗

arXiv2022

Functional Calibration under Non-Probability Survey Sampling

Non-probability sampling is prevailing in survey sampling, but ignoring its selection bias leads to erroneous inferences. We offer a unified nonparametric calibration method to estimate the sampling weights for a non-probability sample by calibrating functions of auxiliary variables in a reproducing kernel Hilbert space. The consistency and the limiting distribution of the proposed estimator are established, and the corresponding variance estimator is also investigated. Compared with existing works, the proposed method is more robust since no parametric assumption is made for the selection mechanism of the non-probability sample. Numerical results demonstrate that the proposed method outperforms its competitors, especially when the model is misspecified. The proposed method is applied to analyze the average total cholesterol of Korean citizens based on a non-probability sample from the National Health Insurance Sharing Service and a reference probability sample from the Korea National Health and Nutrition Examination Survey.

stat.ME↗

arXiv2022

Nonparametric Feature Selection by Random Forests and Deep Neural Networks

Random forests are a widely used machine learning algorithm, but their computational efficiency is undermined when applied to large-scale datasets with numerous instances and useless features. Herein, we propose a nonparametric feature selection algorithm that incorporates random forests and deep neural networks, and its theoretical properties are also investigated under regularity conditions. Using different synthetic models and a real-world example, we demonstrate the advantage of the proposed algorithm over other alternatives in terms of identifying useful features, avoiding useless ones, and the computation efficiency. Although the algorithm is proposed using standard random forests, it can be widely adapted to other machine learning algorithms, as long as features can be sorted accordingly.

cs.LG↗

arXiv2021

Hypotheses Testing from Complex Survey Data Using Bootstrap Weights: A Unified Approach

Standard statistical methods that do not take proper account of the complexity of survey design can lead to erroneous inferences when applied to survey data due to unequal selection probabilities, clustering, and other design features. In particular, the actual type I error rates of tests of hypotheses using standard methods can be much bigger than the nominal significance level. Methods that take account of survey design features in testing hypotheses have been proposed, including Wald tests and quasi-score tests that involve the estimated covariance matrices of parameter estimates. In this paper, we present a unified approach to hypothesis testing that does not require computing the covariance matrices by constructing bootstrap approximations to weighted likelihood ratio statistics and weighted quasi-score statistics and establish the asymptotic validity of the proposed bootstrap tests. In addition, we also consider hypothesis testing from categorical data and present a bootstrap procedure for testing simple goodness of fit and independence in a two-way table. In the simulation studies, the type I error rates of the proposed approach are much closer to their nominal significance level compared with the naive likelihood-ratio test and quasi-score test. An application to data from an educational survey under a logistic regression model is also presented.

stat.ME↗

arXiv2019

Matrix Completion for Survey Data Prediction with Multivariate Missingness

The National Health and Nutrition Examination Survey (NHANES) studies the nutritional and health status over the whole U.S. population with comprehensive physical examinations and questionnaires. However, survey data analyses become challenging due to inevitable missingness in almost all variables. In this paper, we develop a new imputation method to deal with multivariate missingness at random using matrix completion. In contrast to existing imputation schemes either conducting row-wise or column-wise imputation, we treat the data matrix as a whole which allows exploiting both row and column patterns to impute the missing values in the whole data matrix at one time. We adopt a column-space-decomposition model for the population data matrix with easy-to-obtain demographic data as covariates and a low-rank structured residual matrix. A unique challenge arises due to lack of identification of parameters in the sample data matrix. We propose a projection strategy to uniquely identify the parameters and corresponding penalized estimators, which are computationally efficient and possess desired statistical properties. The simulation study shows that the doubly robust estimator using the proposed matrix completion for imputation has smaller mean squared error than other competitors. To demonstrate practical relevance, we apply the proposed method to the 2015-2016 NHANES Questionnaire Data.

stat.ME↗

arXiv2019

Bootstrap inference for the finite population total under complex sampling designs

Bootstrap is a useful tool for making statistical inference, but it may provide erroneous results under complex survey sampling. Most studies about bootstrap-based inference are developed under simple random sampling and stratified random sampling. In this paper, we propose a unified bootstrap method applicable to some complex sampling designs, including Poisson sampling and probability-proportional-to-size sampling. Two main features of the proposed bootstrap method are that studentization is used to make inference, and the finite population is bootstrapped based on a multinomial distribution by incorporating the sampling information. We show that the proposed bootstrap method is second-order accurate using the Edgeworth expansion. Two simulation studies are conducted to compare the proposed bootstrap method with the Wald-type method, which is widely used in survey sampling. Results show that the proposed bootstrap method is better in terms of coverage rate especially when sample size is limited.

math.ST↗

arXiv2018

Sampling techniques for big data analysis in finite population inference

In analyzing big data for finite population inference, it is critical to adjust for the selection bias in the big data. In this paper, we propose two methods of reducing the selection bias associated with the big data sample. The first method uses a version of inverse sampling by incorporating auxiliary information from external sources, and the second one borrows the idea of data integration by combining the big data sample with an independent probability sample. Two simulation studies show that the proposed methods are unbiased and have better coverage rates than their alternatives. In addition, the proposed methods are easy to implement in practice.

stat.ME↗