Title:
How Robust Are Canonical Experimental Economics Results?
How Robust Are Canonical Experimental Economics Results?
Speaker:
林伯軒 (Po-Hsuan Lin)
林伯軒 (Po-Hsuan Lin)
Time:
05/20 (Sun.) 6 pm PDT, 7 pm MDT, 8 pm CDT, 9 pm EDT
05/21 (Mon.) 9 am Taiwan
*注意：本週演講往後調整到週日 (US) / 週一 (TW)
05/20 (Sun.) 6 pm PDT, 7 pm MDT, 8 pm CDT, 9 pm EDT
05/21 (Mon.) 9 am Taiwan
Keywords:
Economics, Experimental Economics, Experimental Methodology
Economics, Experimental Economics, Experimental Methodology
Abstract:
Conventional wisdom suggests that experimental results due to other regarding preferences may vary greatly across cultures, while results that concern markets institutions will remain constant. However, existing data that supports such conclusion is sparse; it is often cost-prohibitive and not pragmatic to run the same experiment across a large number of countries. Thus, we are either left to compare results of slight variations of designs across many countries or the same design across very few countries. Our paper provides an alternative. Utilizing data from MobLab educational platform, we are able to examine data from 10 different countries to test whether the players’ behavior is consistent with theoretical prediction across different countries. We focus on results of identical experiments of the well-known Ultimatum Game and Double Auction.
Conventional wisdom suggests that experimental results due to other regarding preferences may vary greatly across cultures, while results that concern markets institutions will remain constant. However, existing data that supports such conclusion is sparse; it is often cost-prohibitive and not pragmatic to run the same experiment across a large number of countries. Thus, we are either left to compare results of slight variations of designs across many countries or the same design across very few countries. Our paper provides an alternative. Utilizing data from MobLab educational platform, we are able to examine data from 10 different countries to test whether the players’ behavior is consistent with theoretical prediction across different countries. We focus on results of identical experiments of the well-known Ultimatum Game and Double Auction.
Results generally support conventional wisdom. In ultimatum game data, we note a significant country heterogeneity in proposer offers but not responders' behavior. We also note subtle effects of 50/50 offers on responder's behavior both in response and decision time previously not reported in studies with smaller sample time. In the double auction, we find little differences in “mean error deviation”, “Smith’s alpha”, and “efficiency”, three commonly calculated measures of market convergence.
No comments:
Post a Comment