登入選單
返回Google圖書搜尋
Numbers for Pascal - Author Effects in a Meta-Analysis of Global Trade Liberalization Simulations
註釋We perform a meta-analysis of 110 studies that report simulated global welfare gains from trade liberalization under the WTO Doha Development Agenda (DDA) to test for hidden, author-specific model specification effects. We find that after controlling for other factors that influence simulation results, author fixed effects account for more than 10% of the total variation in simulated welfare gains, and up to one-third of the achievable R2. It is not the lead author of a paper who introduces the strongest subjective component, but rather that co-author who has contributed to the most other papers in our literature sample. Hence, our results suggest that empirical simulations of DDA trade liberalization outcomes depend heavily on the subjective decisions of experienced senior researchers who are associated with specific modeling approaches.