How Digital Trade Barriers Affect Cross-border E-commerce
Research design and staged development using multi-source panel data
- Period
- 2025 — Ongoing
- Status
- Ongoing
- Type
- Research project
- My role
- Research question design · Literature review · Data-source mapping · Variable design · Model design · Academic writing
Overview
An ongoing study of the relationship between data-localization policies, digital trade restrictions and cross-border e-commerce using panel data and a two-way fixed-effects design.
Background
Digital trade rules can affect data flows, platform operations and transaction costs. The study plans to combine the OECD Digital STRI, World Bank macroeconomic data and CEPII distance data into a testable panel-data structure.
Core questions
- 01How can digital trade restrictions be translated into comparable measures?
- 02How should sources with different coverage, frequency and definitions be aligned?
- 03Which variables and model structure can account for country and year effects?
- 04Which robustness and mechanism checks are needed?
My responsibilities
- Defined the research question and analytical boundaries.
- Mapped data sources from the OECD, World Bank and CEPII.
- Designed outcome, explanatory and control variables and the model structure.
- Reviewed literature and developed the research and paper outline.
Process
Research
Reviewed work on digital trade barriers, data localization and cross-border commerce.
Design
Planned the indicators, variables, panel structure and two-way fixed-effects model.
Delivery
Advanced definition checks, source matching and research documentation.
Review
Planned robustness, alternative-variable and mechanism tests; no publishable final result is available yet.
Output
Produced the research framework, source inventory and staged paper structure.
Outcomes
- Research question and hypothesis framework
- Multi-source data inventory
- Variable and model design
- Staged paper structure
Evidence & materials
This work is at the research-design and staged-development phase. No unvalidated final conclusion is presented, and the full dataset and paper are not public.
Limitations & reflection
Comparability and transparent definitions across sources determine the credibility of the analysis. The next phase is data cleaning, model testing and result review.