Qinyi Long
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qinyi Long has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
Dong Jiang
School of Statistics and Data Science, Zhejiang Gongshang University
2 shared publications- Dong Jiang
Zhejiang Gongshang University
2 shared publications - Bo Peng
Department of Economics and Management, Bozhou University, Bozhou, China
1 shared publication - Kai Yuan
Bozhou University, Bozhou, Anhui, China
1 shared publication - Xiaojun Cheng1 shared publication
- Zhengsong Xu
Department of Economics and Management, Bozhou University, Bozhou, China
1 shared publication
Latest Publications
- 10 November 2026
- Published by IOP Publishing in Physica Scripta
- Vol. 101 (45) , 455206
Optimal reliability test design for the Chen lifetime distribution under generalized Type-III hybrid censoring
This paper develops optimal test designs for the Chen lifetime distribution under generalized Type-III hybrid censoring, whose four parameters jointly determine the termination rule according to the observed early-failure process. Profile maximum likelihood estimation and the observed information matrix are derived for the censored sample. The asymptotic variance of mission-time reliability is adopted as the local c-optimality criterion, and two failure thresholds and two inspection times are jointly selected subject to constraints on expected cost, expected test duration, minimum failure information, and the upper-tail risk of test duration. To account for uncertainty in the pilot parameter estimates, Bayesian and standardized minimax-regret designs are further developed. Simulation results show that, under the prescribed resource constraints, the generalized Type-III scheme achieves reliability-estimation precision comparable to or better than that of existing schemes. The improvement is more pronounced in the scenario with a rapidly increasing hazard rate considered in this study. Under parameter setting B , the local, Bayesian, and standardized minimax-regret criteria select the same plan. A rolling-bearing lifetime dataset is used to illustrate how a follow-up test plan can be determined from pilot estimates.
- 10 November 2026
- Published by IOP Publishing in Physica Scripta
- Vol. 101 (45) , 455206
Optimal reliability test design for the Chen lifetime distribution under generalized Type-III hybrid censoring
This paper develops optimal test designs for the Chen lifetime distribution under generalized Type-III hybrid censoring, whose four parameters jointly determine the termination rule according to the observed early-failure process. Profile maximum likelihood estimation and the observed information matrix are derived for the censored sample. The asymptotic variance of mission-time reliability is adopted as the local c-optimality criterion, and two failure thresholds and two inspection times are jointly selected subject to constraints on expected cost, expected test duration, minimum failure information, and the upper-tail risk of test duration. To account for uncertainty in the pilot parameter estimates, Bayesian and standardized minimax-regret designs are further developed. Simulation results show that, under the prescribed resource constraints, the generalized Type-III scheme achieves reliability-estimation precision comparable to or better than that of existing schemes. The improvement is more pronounced in the scenario with a rapidly increasing hazard rate considered in this study. Under parameter setting B , the local, Bayesian, and standardized minimax-regret criteria select the same plan. A rolling-bearing lifetime dataset is used to illustrate how a follow-up test plan can be determined from pilot estimates.
- 1 September 2026
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- p. 1-21
Estimation and prediction of the inverted exponentiated Rayleigh distribution under adaptive progressive double Type-II hybrid censored data
1. The study of censored data originated from survival analysis and reliability engineering. In practical applications, due to limitations in experimental time, cost, or observational constraints, ...
- 1 September 2026
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- p. 1-21
Estimation and prediction of the inverted exponentiated Rayleigh distribution under adaptive progressive double Type-II hybrid censored data
1. The study of censored data originated from survival analysis and reliability engineering. In practical applications, due to limitations in experimental time, cost, or observational constraints, ...
- Open Access
- 25 July 2026
- Published by Elsevier in SSRN Electronic Journal
MAPatchTST: Local-Structure-Aware Patch Encoding and Graph-Clustering-Guided Hybrid Channel Modeling for Multivariate Time Series Forecasting
Patch-based Transformers provide an effective framework for long-horizon multivariate time series forecasting by reducing the effective sequence length and preserving local continuity. However, conventional patch tokenization usually projects each segment as a static block, which may weaken trend transitions and short-term deviations within the patch. Channel-independent forecasting reduces cross-variable interference but discards useful dependencies, whereas fully channel-dependent modeling can propagate weak correlations and noise across variables. To address these limitations, this paper proposes MAPatchTST, which jointly improves fixed-scale patch representation and channel interaction. A mean-adaptive patch encoder first decomposes each patch into a running local baseline and a deviation component. Learnable temporal decay, deviation modulation, and intra-patch attention are then applied to preserve local changes before global temporal encoding. Based on the enhanced patch representations, a sparse association graph is constructed to generate task-oriented channel embeddings. The channels are subsequently clustered according to their propagated representations. Dependency modeling is performed within each cluster, while cross-cluster information exchange is restricted to reduce noise propagation. A residual prediction objective with regularization is further adopted to stabilize model training under non-stationary conditions. Experiments on ETTh1, ETTh2, ETTm1, ETTm2, Electricity, and Traffic show that MAPatchTST consistently improves the direct PatchTST baseline, reducing average MSE and MAE by approximately 3.55% and 4.05%, respectively. Ablation and sensitivity analyses further demonstrate the distinct contributions of local patch enhancement, sparse channel organization, and cluster-wise interaction.
- 25 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMAPatchTST: Local-Structure-Aware Patch Encoding and Graph-Clustering-Guided Hybrid Channel Modeling for Multivariate Time Series Forecasting
Patch-based Transformers provide an effective framework for long-horizon multivariate time series forecasting by reducing the effective sequence length and preserving local continuity. However, conventional patch tokenization usually projects each segment as a static block, which may weaken trend transitions and short-term deviations within the patch. Channel-independent forecasting reduces cross-variable interference but discards useful dependencies, whereas fully channel-dependent modeling can propagate weak correlations and noise across variables. To address these limitations, this paper proposes MAPatchTST, which jointly improves fixed-scale patch representation and channel interaction. A mean-adaptive patch encoder first decomposes each patch into a running local baseline and a deviation component. Learnable temporal decay, deviation modulation, and intra-patch attention are then applied to preserve local changes before global temporal encoding. Based on the enhanced patch representations, a sparse association graph is constructed to generate task-oriented channel embeddings. The channels are subsequently clustered according to their propagated representations. Dependency modeling is performed within each cluster, while cross-cluster information exchange is restricted to reduce noise propagation. A residual prediction objective with regularization is further adopted to stabilize model training under non-stationary conditions. Experiments on ETTh1, ETTh2, ETTm1, ETTm2, Electricity, and Traffic show that MAPatchTST consistently improves the direct PatchTST baseline, reducing average MSE and MAE by approximately 3.55% and 4.05%, respectively. Ablation and sensitivity analyses further demonstrate the distinct contributions of local patch enhancement, sparse channel organization, and cluster-wise interaction.
- 1 September 2025
- Published by IOP Publishing in Physica Scripta
- Vol. 100 (9) , 095240
Parameter estimation and reliability analysis of Burr XII distribution under generalized type-II censoring
A generalized type-II censoring scheme is proposed based on the traditional type-II censoring scheme. Maximum likelihood estimators of the parameters are developed and approximate confidence intervals are constructed for generalized type-II censored sample and Burr XII distribution. The bootstrap methods are used to derive two types of confidence intervals for the parameters. Under the squared error loss function, Bayesian estimates of the parameters are obtained using Lindley’s approximation and Markov chain Monte Carlo (MCMC) methods. In addition, the estimation of reliability for series and parallel systems is discussed. Comparison through Monte Carlo simulations reveals that Bayesian estimation significantly outperforms maximum likelihood estimation under the mean squared error criterion. Finally, the validity and practicability of the proposed methods is verified through actual data analysis.
- 1 September 2025
- Published by IOP Publishing in Physica Scripta
- Vol. 100 (9) , 095240
Parameter estimation and reliability analysis of Burr XII distribution under generalized type-II censoring
A generalized type-II censoring scheme is proposed based on the traditional type-II censoring scheme. Maximum likelihood estimators of the parameters are developed and approximate confidence intervals are constructed for generalized type-II censored sample and Burr XII distribution. The bootstrap methods are used to derive two types of confidence intervals for the parameters. Under the squared error loss function, Bayesian estimates of the parameters are obtained using Lindley’s approximation and Markov chain Monte Carlo (MCMC) methods. In addition, the estimation of reliability for series and parallel systems is discussed. Comparison through Monte Carlo simulations reveals that Bayesian estimation significantly outperforms maximum likelihood estimation under the mean squared error criterion. Finally, the validity and practicability of the proposed methods is verified through actual data analysis.