A5020490822- Stock Market Forecasting Methods
- Complex Network Analysis Techniques
- Energy Load and Power Forecasting
- Advanced Text Analysis Techniques
- Opinion Dynamics and Social Influence
- Hydrocarbon exploration and reservoir analysis
- Reservoir Engineering and Simulation Methods
- Caching and Content Delivery
- Grey System Theory Applications
- Market Dynamics and Volatility
- Atmospheric and Environmental Gas Dynamics
- Network Security and Intrusion Detection
- Energy, Environment, Economic Growth
- Financial Markets and Investment Strategies
- Innovation Policy and R&D
- Financial Risk and Volatility Modeling
Communication University of China
2023-2024
Chongqing Technology and Business University
2022-2024
Southwestern University of Finance and Economics
2024
Multinomial logit model(MLM) has been proposed as the most frequently regression model for multi-category response. To deal with correlated data, in this paper we propose G-LASSO/G-SCAD/G-MCP penalized MLM to exert class discovery and prediction high-dimensional classification problems. Firstly, develop a group coordinate descent(GCD) algorithm simultaneously complete selection estimation, prove convergence GCD under mild conditions. Secondly, apply training set estimations obtain...
Correctly predicting the stock price movement direction is of immense importance in financial market. In recent years, with expansion dimension and volume data, nonstationary nonlinear characters finance data make it difficult to predict accurately. this article, we propose a methodology that combines technical analysis sentiment construct predictor variables then apply improved LASSO-LASSO forecast direction. First, textual content historical transaction are crawled from websites. Then...
As a clean unconventional energy source, shale gas reservoirs are increasingly important globally. Accurate prediction methods for production capacity can bring significant economic benefits by reducing construction and operating costs. Decline curve analysis (DCA) is an efficient method that uses mathematical formulas to describe trends with minimal reliance on geological or engineering parameters. However, traditional DCA models often fail capture the complex dynamics of wells, especially...
Abstract Regional innovation output is influenced by many factors such as macroeconomic environments, residents consumption, fixed asset investment, foreign trade, fiscal revenue and expenditure, education, research development (R&D) input. Correctly predicting regional an important subject in the economic field. In this paper, we propose four regularized Poisson regressions to forecast for 31 provinces China. Firstly, screen out 20 combine with penalties: ridge penalty , least absolute...
In recent years, machine learning and data mining methods have been applied to stock price forecasting problems. However, due the time series characteristics, noise multicollinearity of data, many traditional cannot accurately predict trend movement. This paper applies Least Absolute Shrinkage Selection Operator (LASSO) penalized logistic regression model ridge prediction problem evaluates their performance. First, we select CMA, EQR IRM from 2016-07-21 2020-07-13. Then apply TTR package...
Identifying super-propagators on social networks takes critical significance in guiding and controlling public opinions the internet. The influence of nodes propagation process not only depends network structure, but also is related to dynamic interaction between nodes. To address this problem, from perspectives topology interaction, a multi-factor information matrix centrality algorithm developed study accordance with characteristics three degrees rule, node influence, neighbor mutual...
In respect of social networks, the identification pivotal nodes plays a role in revealing inherent characteristics and overall network structure, for which it is applicable to serve various purposes such as enhancing propagation information, reducing misinformation, conducting precision marketing campaigns. Allowing directed nature proposed this study introduce an improved method based on gravity centrality. Firstly, we incorporate node's reachability global positional characteristics. These...