arXiv subjects

Qianqian Zhu

Publications and source records attributed to Qianqian Zhu.

At least 19 recordsLinked to original sources

arXiv2026

Mixed-Frequency Time Series Forecasting via Depth-Separable Neural Networks

To better forecast mixed-frequency time series, it is the key to choose a suitable way for frequency alignment. However, the existing methods are all limited to linear transformations, and this may overlook the possible nonlinearity, leading to a worse prediction. We alternatively consider a deep neural network for each frequency alignment, and hence a depth-separable neural network. Moreover, a parameter-sharing mechanism is adopted across the alignment at each stage, making possible a deeper network for a large set of higher-frequency predictors. This paper establishes an approximation theory for the proposed depth-separable network, and a non-asymptotic prediction error bound is also derived. Simulation studies demonstrate the finite-sample performance of the proposed method, and an empirical application to forecasting U.S. quarterly macroeconomic variables using monthly and daily indicators, highlights its superior predictive accuracy over existing mixed-frequency methods.

stat.ME↗

arXiv2026

A computationally efficient multivariate volatility model for many assets

This paper develops a flexible and computationally efficient multivariate volatility model that accommodates dynamic conditional correlations and volatility spillover effects among financial assets. The new model has desirable properties such as identifiability and computational tractability for many assets.A sufficient condition for strict stationarity is derived for the process.Two quasi-maximum likelihood estimation methods are proposed for the new model without and with low-rank constraints on the coefficient matrices, respectively, and the asymptotic properties of both estimators are established. Moreover, a selection-consistent Bayesian information criterion is developed for order selection, and testing for volatility spillover effects is discussed. The finite-sample performance of the proposed methodology is evaluated in simulations for small and moderate dimensions.Its usefulness and inference tools are illustrated by two empirical examples for 5 stock markets and 17 industry portfolios.

stat.ME↗

arXiv2026

A robust and scalable estimation for high-dimensional volatility models

This paper introduces a robust and computationally efficient estimation framework for high-dimensional volatility models in the BEKK-ARCH class. The proposed approach employs data truncation to ensure robustness against heavy-tailed distributions and utilizes a regularized least squares method for efficient optimization in high-dimensional settings. Non-asymptotic error bounds are established for the resulting estimators under heavy-tailed regimes, and the minimax optimal convergence rate is derived. Moreover, a robust BIC and a Ridge-type estimator are introduced for selecting the model order and the number of BEKK components, respectively, with their selection consistency established under heavy-tailed settings. Simulation studies demonstrate finite-sample performance of the proposed method, and two empirical applications illustrate its practical utility. The results show that the new framework outperforms existing alternatives in both computational speed and forecasting accuracy.

math.ST↗

arXiv2026

Personalized Federated Learning for Tensor Regression

The growing availability of tensor-valued data across multiple institutions creates opportunities for collaborative analysis, but also raises challenges related to data privacy, high dimensionality, and client heterogeneity. This paper introduces a personalized federated tensor regression framework that addresses all three simultaneously. Each client's coefficient tensor is decomposed into a globally shared low-Tucker-rank component and a locally sparse deviation, estimated via a two-stage privacy-preserving procedure. We establish finite-sample upper bounds and minimax lower bounds that quantify the privacy-accuracy trade-off, and prove the consistency of the supporting initialization and rank-selection steps. Simulation studies confirm that the federated approach improves estimation and prediction over purely local methods, especially when per-client data are scarce, and an MRI-based ADHD study illustrates its strong performance under real privacy constraints.

stat.ME↗

arXiv2026

Private Federated Learning for High-dimensional Time Series

In the era of big data, leveraging information from multiple clients while preserving data privacy has emerged as a critical challenge in modern statistical modeling and forecasting. This paper introduces a privacy-preserving federated learning framework for high-dimensional vector autoregressive models, where each client's dynamics are characterized by a common low-rank structure augmented with sparse client-specific deviations. We develop a two-stage estimation procedure that integrates differentially private representation learning for the shared component with local personalization for client-specific adjustments, enabling effective information pooling under selective privacy constraints. Non-asymptotic error bounds are established for both the single-client and federated estimators to characterize the inherent privacy-utility trade-off, and consistency of a ridge-type rank selection criterion is proved. Simulation studies demonstrate that federation substantially improves estimation accuracy when local sample sizes are limited. Two empirical applications to analyzing electricity-economy linkages across U.S. states and conducting multi-task macroeconomic forecasting across countries, highlight the superior predictive accuracy of the proposed method over existing single-client benchmarks.

stat.ME↗

arXiv2026

Quasi-maximum likelihood estimation for scalable ARMA models

The recently proposed scalable ARMA model preserves the parsimony of traditional VARMA models while achieving greater computational tractability. However, existing studies are limited to regularized least squares estimation (LSE) for high-dimensional settings, which is not only statistically less efficient but also requires the sub-Gaussian assumption for its theoretical guarantees. Moreover, it still lacks inference tool for real applications. To fill this gap, we develop a quasi-maximum likelihood estimation (QMLE) framework for scalable ARMA models. Its asymptotic normality is established under a finite fourth order moment condition, and we formally prove its asymptotic efficiency gain over LSE. We also introduce an efficient block coordinate descent algorithm for computation and a consistent Bayesian information criterion for model selection. Simulation studies validate the finite-sample performance of our methodology, and an empirical application to six macroeconomic indicators demonstrates its practical utility.

stat.ME↗

arXiv2025

Improving time series estimation and prediction via transfer learning

There are many time series in the literature with high dimension yet limited sample sizes, such as macroeconomic variables, and it is almost impossible to obtain efficient estimation and accurate prediction by using the corresponding datasets themselves. This paper fills the gap by introducing a novel representation-based transfer learning framework for vector autoregressive models, and information from related source datasets with rich observations can be leveraged to enhance estimation efficiency through representation learning. A two-stage regularized estimation procedure is proposed with well established non-asymptotic properties, and algorithms with alternating updates are suggested to search for the estimates. Our transfer learning framework can handle time series with varying sample sizes and asynchronous starting and/or ending time points, thereby offering remarkable flexibility in integrating information from diverse datasets. Simulation experiments are conducted to evaluate the finite-sample performance of the proposed methodology, and its usefulness is demonstrated by an empirical analysis on 20 macroeconomic variables from Japan and another nine countries.

stat.ME↗

arXiv2025

A Deep Learning System for Rapid and Accurate Warning of Acute Aortic Syndrome on Non-contrast CT in China

The accurate and timely diagnosis of acute aortic syndromes (AAS) in patients presenting with acute chest pain remains a clinical challenge. Aortic CT angiography (CTA) is the imaging protocol of choice in patients with suspected AAS. However, due to economic and workflow constraints in China, the majority of suspected patients initially undergo non-contrast CT as the initial imaging testing, and CTA is reserved for those at higher risk. In this work, we present an artificial intelligence-based warning system, iAorta, using non-contrast CT for AAS identification in China, which demonstrates remarkably high accuracy and provides clinicians with interpretable warnings. iAorta was evaluated through a comprehensive step-wise study. In the multi-center retrospective study (n = 20,750), iAorta achieved a mean area under the receiver operating curve (AUC) of 0.958 (95% CI 0.950-0.967). In the large-scale real-world study (n = 137,525), iAorta demonstrated consistently high performance across various non-contrast CT protocols, achieving a sensitivity of 0.913-0.942 and a specificity of 0.991-0.993. In the prospective comparative study (n = 13,846), iAorta demonstrated the capability to significantly shorten the time to correct diagnostic pathway. For the prospective pilot deployment that we conducted, iAorta correctly identified 21 out of 22 patients with AAS among 15,584 consecutive patients presenting with acute chest pain and under non-contrast CT protocol in the emergency department (ED) and enabled the average diagnostic time of these 21 AAS positive patients to be 102.1 (75-133) mins. Last, the iAorta can help avoid delayed or missed diagnosis of AAS in settings where non-contrast CT remains the unavoidable the initial or only imaging test in resource-constrained regions and in patients who cannot or did not receive intravenous contrast.

eess.IV↗

arXiv2024

AResNet-ViT: A Hybrid CNN-Transformer Network for Benign and Malignant Breast Nodule Classification in Ultrasound Images

To address the challenges of similarity between lesions and surrounding tissues, overlapping appearances of partially benign and malignant nodules, and difficulty in classification, a deep learning network that integrates CNN and Transformer is proposed for the classification of benign and malignant breast lesions in ultrasound images. This network adopts a dual-branch architecture for local-global feature extraction, making full use of the advantages of CNN in extracting local features and the ability of ViT to extract global features to enhance the network's feature extraction capabilities for breast nodules. The local feature extraction branch employs a residual network with multiple attention-guided modules, which can effectively capture the local details and texture features of breast nodules, enhance sensitivity to subtle changes within the nodules, and thus can aid in accurate classification of their benign and malignancy. The global feature extraction branch utilizes the multi-head self-attention ViT network, which can capture the overall shape, boundary, and relationship with surrounding tissues, and thereby enhancing the understanding and modeling of both nodule and global image features. Experimental results on a public ultrasound breast nodule data set show that the proposed method is better than other comparison networks, This indicates that the fusion of CNN and Transformer networks can effectively improve the performance of the classification model and provide a powerful solution for the benign-malignant classification of ultrasound breast.

eess.IV↗

arXiv2024

Quasi-maximum Likelihood Inference for Linear Double Autoregressive Models

This paper investigates the quasi-maximum likelihood inference including estimation, model selection and diagnostic checking for linear double autoregressive (DAR) models, where all asymptotic properties are established under only fractional moment of the observed process. We propose a Gaussian quasi-maximum likelihood estimator (G-QMLE) and an exponential quasi-maximum likelihood estimator (E-QMLE) for the linear DAR model, and establish the consistency and asymptotic normality for both estimators. Based on the G-QMLE and E-QMLE, two Bayesian information criteria are proposed for model selection, and two mixed portmanteau tests are constructed to check the adequacy of fitted models. Moreover, we compare the proposed G-QMLE and E-QMLE with the existing doubly weighted quantile regression estimator in terms of the asymptotic efficiency and numerical performance. Simulation studies illustrate the finite-sample performance of the proposed inference tools, and a real example on the Bitcoin return series shows the usefulness of the proposed inference tools.

stat.ME↗

arXiv2023

Quantile autoregressive conditional heteroscedasticity

This paper proposes a novel conditional heteroscedastic time series model by applying the framework of quantile regression processes to the ARCH(\infty) form of the GARCH model. This model can provide varying structures for conditional quantiles of the time series across different quantile levels, while including the commonly used GARCH model as a special case. The strict stationarity of the model is discussed. For robustness against heavy-tailed distributions, a self-weighted quantile regression (QR) estimator is proposed. While QR performs satisfactorily at intermediate quantile levels, its accuracy deteriorates at high quantile levels due to data scarcity. As a remedy, a self-weighted composite quantile regression (CQR) estimator is further introduced and, based on an approximate GARCH model with a flexible Tukey-lambda distribution for the innovations, we can extrapolate the high quantile levels by borrowing information from intermediate ones. Asymptotic properties for the proposed estimators are established. Simulation experiments are carried out to access the finite sample performance of the proposed methods, and an empirical example is presented to illustrate the usefulness of the new model.

stat.ME↗

arXiv2023

Multichannel consecutive data cross-extraction with 1DCNN-attention for diagnosis of power transformer

Power transformer plays a critical role in grid infrastructure, and its diagnosis is paramount for maintaining stable operation. However, the current methods for transformer diagnosis focus on discrete dissolved gas analysis, neglecting deep feature extraction of multichannel consecutive data. The unutilized sequential data contains the significant temporal information reflecting the transformer condition. In light of this, the structure of multichannel consecutive data cross-extraction (MCDC) is proposed in this article in order to comprehensively exploit the intrinsic characteristic and evaluate the states of transformer. Moreover, for the better accommodation in scenario of transformer diagnosis, one dimensional convolution neural network attention (1DCNN-attention) mechanism is introduced and offers a more efficient solution given the simplified spatial complexity. Finally, the effectiveness of MCDC and the superior generalization ability, compared with other algorithms, are validated in experiments conducted on a dataset collected from real operation cases of power transformer. Additionally, the better stability of 1DCNN-attention has also been certified.

cs.LG↗

arXiv2021

Asymmetric linear double autoregression

This paper proposes the asymmetric linear double autoregression, which jointly models the conditional mean and conditional heteroscedasticity characterized by asymmetric effects. A sufficient condition is established for the existence of a strictly stationary solution. With a quasi-maximum likelihood estimation (QMLE) procedure introduced, a Bayesian information criterion (BIC) and its modified version are proposed for model selection. To detect asymmetric effects in the volatility, the Wald, Lagrange multiplier and quasi-likelihood ratio test statistics are put forward, and their limiting distributions are established under both null and local alternative hypotheses. Moreover, a mixed portmanteau test is constructed to check the adequacy of the fitted model. All asymptotic properties of inference tools including QMLE, BICs, asymmetric tests and the mixed portmanteau test, are established without any moment condition on the data process, which makes the new model and its inference tools applicable for heavy-tailed data. Simulation studies indicate that the proposed methods perform well in finite samples, and an empirical application to S\&P500 Index illustrates the usefulness of the new model.

stat.ME↗

arXiv2020

Quantile double autoregression

Many financial time series have varying structures at different quantile levels, and also exhibit the phenomenon of conditional heteroscedasticity at the same time. In the meanwhile, it is still lack of a time series model to accommodate both of the above features simultaneously. This paper fills the gap by proposing a novel conditional heteroscedastic model, which is called the quantile double autoregression. The strict stationarity of the new model is derived, and a self-weighted conditional quantile estimation is suggested. Two promising properties of the original double autoregressive model are shown to be preserved. Based on the quantile autocorrelation function and self-weighting concept, two portmanteau tests are constructed, and they can be used in conjunction to check the adequacy of fitted conditional quantiles. The finite-sample performance of the proposed inference tools is examined by simulation studies, and the necessity of the new model is further demonstrated by analyzing the S&P500 Index.

math.ST↗

arXiv2019

Statistical analysis on X-ray flares from the nucleus and HST-1 knot in the M87 jet

The statistical properties of X-ray flares from two separate locations (nucleus and HST-1) in the M87 jet are investigated to reveal the physical origin of the flares. We analyse the archival \textit{Chandra} data for M87, and identify 14 flares in the nucleus and 9 flares in HST-1. The peak intensity ($I_{\rm{P}}$) and the flaring duration time ($T_{\rm{fl}}$) for each flare are obtained. It is found that the distributions of both $I_{\rm{P}}$ and $T_{\rm{fl}}$ for the nucleus obey a power-law form with a similar index. A similar result is also obtained for HST-1, and no significant inconsistency between the nucleus and HST-1 is found for the indices. Similar to solar X-ray flares, the power-law distributions of the flare event parameters can be well explained by a self-organized criticality (SOC) system, which are triggered by magnetic reconnection. Our results suggest that the flares from nucleus and HST-1 are possibly triggered by magnetic reconnection process. The consistent indices for the distributions of $I_{\rm{P}}$ and $T_{\rm{fl}}$ in the CORE and HST-1 indicate that the dimensions of the energy dissipation of the magnetic reconnection is identical in the two regions. A strong correlation between the flares in the two regions also suggests a similar physical origin for the flares.

astro-ph.HE↗

arXiv2018

Impact of Plasma Instability on Constraint of the Intergalactic Magnetic Field

A relativistic electron-positron pair beam can be produced in the interaction of TeV photons from a blazar with the extragalactic background light (EBL). The relativistic $e^{\pm}$ pairs would loss energy through inverse-Compton scattering (ICS) photons of cosmic microwave background (CMB) or plasma instabilities. The dominant energy-loss process is under debate. Based on the assumption that the dominant energy-loss process is ICS, the resulted cascade GeV radiation is usually used to constrain the intergalactic magnetic field (IGMF). Here, we include the energy-loss due to plasma oblique instability in the calculation of cascade gamma-ray flux, and investigate the impact of the plasma instability on the constraint of IGMF. The up-to-date GeV data and archival TeV data of the blazar 1ES 0229+200 are used. The results indicate that even if the oblique instability cooling is dominating over ICS cooling, the cascade flux could be still used to constrain the IGMF. It is found that with the ratio between the cooling rates of the oblique instability and the ICS varying from 0.1, 1 to 10, the lower limit of the IGMF putted by the cascade flux and the gamma-ray data changes from $8\times10^{-18}\ $G, $5\times10^{-18}\ $G to $10^{-18}\ $G. If the ratio between the two cooling rates is 30, the estimate of IGMF based on the cascade flux is invalid.

astro-ph.HE↗

arXiv2018

Testing relativistic boost as the cause of gamma-ray quasi-periodic oscillation in a blazar

The mechanism for producing gamma-ray quasi-periodic oscillation (QPO) in blazar is unknown. One possibility is the geometric model, in which without the need for intrinsic quasi-periodic variation, the relativistic Doppler factor changes periodically, resulting in observed gamma-ray QPO. We propose a method to test this geometric model. We analyze the {\it Fermi}-LAT data of PG 1553+113 spanning from 2008 August until 2018 February. According to 29 four-month average spectral energy distributions (SEDs) in the energy range of 0.1-300 GeV, we split the {\it Fermi}-LAT energy range into three bands: 0.1-1 GeV, 1-10 GeV, and 10-300 GeV. The spectrum in each energy range can be successfully fitted by a power-law. The light curves and photon indices in the three energy ranges are obtained. Then, light curves in three narrow energy ranges, i.e., 0.2-0.5 GeV, 2-5 GeV and 20- 40 GeV, are constructed, and the relative variability amplitudes in the three narrow energy ranges are calculated. A discrete-correlation analysis is performed for the light curves. Our results indicate that (i) the light curves in the different energy ranges follow the same pattern showed in the light curve above 0.1 GeV; (ii) the three groups of photon indices in the energy ranges of 0.1-1 GeV, 1-10 GeV, and 10-300 GeV keep nearly constant; (iii) the ratio between relative variability amplitudes in different narrow energy ranges are equal (within their errors) to the prediction by the Doppler effect. Our results support the scenario of the relativistic boost producing the gamma-ray QPO for PG 1553+113.

astro-ph.HE↗

arXiv2016

Hybrid Quantile Regression Estimation for Time Series Models with Conditional Heteroscedasticity

Estimating conditional quantiles of financial time series is essential for risk management and many other applications in finance. It is well-known that financial time series display conditional heteroscedasticity. Among the large number of conditional heteroscedastic models, the generalized autoregressive conditional heteroscedastic (GARCH) process is the most popular and influential one. So far, feasible quantile regression methods for this task have been confined to a variant of the GARCH model, the linear GARCH model, owing to its tractable conditional quantile structure. This paper considers the widely used GARCH model. An easy-to-implement hybrid conditional quantile estimation procedure is developed based on a simple albeit nontrivial transformation. Asymptotic properties of the proposed estimator and statistics are derived, which facilitate corresponding inferences. To approximate the asymptotic distribution of the quantile regression estimator, we introduce a mixed bootstrapping procedure, where a time-consuming optimization is replaced by a sample averaging. Moreover, diagnostic tools based on the residual quantile autocorrelation function are constructed to check the adequacy of the fitted conditional quantiles. Simulation experiments are carried out to assess the finite-sample performance of the proposed approach. The favorable performance of the conditional quantile estimator and the usefulness of the inference tools are further illustrated by an empirical application.

stat.ME↗