A5065747278- Advanced Causal Inference Techniques
- Spatial and Panel Data Analysis
- Superconducting Materials and Applications
- Particle accelerators and beam dynamics
- Magnetic confinement fusion research
- Gyrotron and Vacuum Electronics Research
- demographic modeling and climate adaptation
- Bayesian Modeling and Causal Inference
- Economic Policies and Impacts
- Atomic and Subatomic Physics Research
- Electromagnetic Launch and Propulsion Technology
- Fusion materials and technologies
- Ionosphere and magnetosphere dynamics
- Structural Health Monitoring Techniques
- Ultrasonics and Acoustic Wave Propagation
- Fatigue and fracture mechanics
- Complex Network Analysis Techniques
- Radiation Detection and Scintillator Technologies
Wuhan Institute of Technology
2025
Massachusetts Institute of Technology
2020-2022
Moscow Institute of Thermal Technology
2022
Max Planck Institute for Plasma Physics
2009-2021
Max Planck Society
2012-2021
This paper introduces a simple framework of counterfactual estimation for causal inference with time-series cross-sectional data, in which we estimate the average treatment effect on treated by directly imputing outcomes observations. We discuss several novel estimators under this framework, including fixed effects estimator, interactive and matrix completion estimator. They provide more reliable estimates than conventional twoway models when are heterogeneous or unobserved time-varying...
ASDEX Upgrade was operated with a fully W-covered wall in 2007 and 2008. Stationary H-modes at the ITER target values improved H up to 1.2 were run without any boronization. The boundary conditions set by full W (high enough ELM frequency, high central heating low power density arriving plates) require significant scenario development, but will apply as well. D retention has been reduced stationary operation saturated found. Concerning confinement, impurity ion transport across pedestal is...
We develop an R package panelView and a Stata panelview for panel data visualization. They are designed to assist causal analysis with have three main functionalities: (1) they plot the treatment status missing values in dataset; (2) visualize variables of interest time-series fashion; (3) depict bivariate relationships between variable outcome either by unit or aggregate. These tools can help researchers better understand their before conducting statistical analysis.
First Page