A5004079307- Spatial and Panel Data Analysis
- Statistical Methods and Inference
- Complex Network Analysis Techniques
- Advanced Statistical Methods and Models
- Energy, Environment, Economic Growth
- Monetary Policy and Economic Impact
- Complex Systems and Time Series Analysis
- Advanced oxidation water treatment
- Advanced Photocatalysis Techniques
- Housing Market and Economics
- Regional Economics and Spatial Analysis
- Statistical Methods and Bayesian Inference
- Technical Engine Diagnostics and Monitoring
- Tensor decomposition and applications
- Bayesian Modeling and Causal Inference
- Adsorption, diffusion, and thermodynamic properties of materials
- Nanomaterials for catalytic reactions
- demographic modeling and climate adaptation
- Financial Risk and Volatility Modeling
- Mechanical and Thermal Properties Analysis
- Market Dynamics and Volatility
- Robotic Path Planning Algorithms
- Vehicle Routing Optimization Methods
- Economic Growth and Productivity
- Tunneling and Rock Mechanics
Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou)
2025
University of Southampton
2024
Shanghai University
2023
University of Reading
2022
University of York
2020-2021
North Carolina State University
2019-2021
Hunan University
2016-2020
Hunan University of Finance and Economics
2017
PLA Information Engineering University
2014
We propose a dynamic network quantile regression model to investigate the connectedness using predetermined information. extend existing autoregression of Zhu et al. by explicitly allowing contemporaneous effects and controlling for common factors across quantiles. To cope with endogeneity issue due simultaneous spillovers, we adopt instrumental variable (IVQR) estimation derive consistency asymptotic normality IVQR estimator near epoch dependence property process. Via Monte Carlo...
A multivariate mixture model is determined by three elements: the number of components, mixing proportions, and component distributions. Assuming that components given ...
We study the estimation of nonlinear models with cross-sectional data using two-step generalized estimating equations within quasi-maximum likelihood framework. To improve efficiency, we propose a grouped estimator that accounts for potential spatial correlation in underlying innovations models. Under mild weak dependence assumptions, provide results on consistency and asymptotic normality. Monte Carlo simulations demonstrate efficiency gain our approach compared to various methods. Finally,...
Security Enhanced Android is the integration of with SE Linux launched by NSA to strengthen security.This system adopting Mandatory Access Control prevents attacks and enforces application isolation.Also, it provides an implementation in current environment.Therefore capabilities defeating root overflow deficiency applications are significantly strengthened.In this paper, we discussed security mechanisms Android, introduced difficulties solutions about implementing from kernel user space level.
We propose a dynamic network quantile regression model to investigate thequantile connectedness using predetermined information. extend theexisting autoregression of Zhu et al. (2019b) by explicitly allowingthe contemporaneous effects and controlling for the common factorsacross quantiles. To cope with endogeneity issue due simultaneous networkspillovers, we adopt instrumental variable (IVQR) estimationand derive consistency asymptotic normality IVQR estimatorusing near epoch dependence...
On the one hand, rapid development of e-commerce has put forward higher requirements and standards for logistics industry, which forced transformation terminal delivery model; on other drones have developed fast in field due to their high flexibility efficiency. This paper summarizes domestic foreign research status vehicle route problem four combined distribution modes, points out future direction.
Penalised likelihood methods have been a success in analysing high dimensional data. Tang and Leng [(2010), ‘Penalized High-Dimensional Empirical Likelihood’, Biometrika, 97(4), 905–920] extended the penalisation approach to empirical scenario showed that penalised estimator could identify true predictors consistently linear regression models. However, this desired selection consistency property of method relies heavily on choice tuning parameter. In work, we propose parameter procedure for...
This paper considers the latent Gaussian graphical model, which extends model to handle discrete data as well mixed with both continuous and variables by assuming that are generated discretizing variables. We propose a modified expectation‐maximization (EM) algorithm estimate parameters in for binary data. also extend proposed EM The conditional dependence structure can be consequently constructed exploring sparsity pattern of precision matrix illustrate performance our estimator through...
We propose a dynamic network quantile regression model to investigate the connectedness using predetermined information. extend existing autoregression of Zhu et al. (2019b) by explicitly allowing contemporaneous effects and controlling for common factors across quantiles. To cope with endogeneity issue due simultaneous spillovers, we adopt instrumental variable (IVQR) estimation derive consistency asymptotic normality IVQR estimator near epoch dependence property process. Via Monte Carlo...