Calendrier du 16 septembre 2024
Roy Seminar (ADRES)
Du 16/09/2024 de 16:00 à 17:15
R1-09
BATTIGALLI Pierpalo (Bocconi)
Monotonicity and Robust Implementation Under Forward-Induction Reasoning
It is well known that, in sequential games, the set of paths consistent with rationality and forward-induction reasoning may change non-monotonically with respect to transparent restrictions on players' beliefs (see Battigalli & Siniscalchi J. Econ. Theory 2002, and Battigalli & Friedenberg, Theor. Econ. 2012). Yet, we prove that---in an incomplete-information environment---predictions become sharper when the restrictions only concern initial beliefs about types. This monotonicity theorem implies that strong rationalizability for games with payoff uncertainty characterizes the path-predictions of forward-induction reasoning across all possible restrictions to players' initial hierarchies of beliefs on the exogenous uncertainty. The latter result allows us to solve an open problem in implementation theory: The implementation of social choice functions through sequential mechanisms under forward-induction reasoning---which considerably expands the realm of implementable functions compared with simultaneous-move mechanisms (Mueller, J. Econ. Theory 2016)---is robust in the sense of Bergemann and Morris (Theor. Econ. 2009).
Econometrics Seminar
Du 16/09/2024 de 16:00 à 17:15
ZOOM
CHEN Kevin (SIEPR)
Empirical Bayes When Estimation Precision Predicts Parameters
Empirical Bayes methods usually maintain a prior independence assumption: The unknown parameters of interest are independent from the known standard errors of the estimates. This assumption is often theoretically questionable and empirically rejected. This paper instead models the conditional distribution of the parameter given the standard errors as a flexibly parametrized family of distributions, leading to a family of methods that we call CLOSE. This paper establishes that (i) CLOSE is rate-optimal for squared error Bayes regret, (ii) squared error regret control is sufficient for an important class of economic decision problems, and (iii) CLOSE is worst-case robust when our assumption on the conditional distribution is misspecified. Empirically, using CLOSE leads to sizable gains for selecting high-mobility Census tracts. Census tracts selected by CLOSE are substantially more mobile on average than those selected by the standard shrinkage method.
Régulation et Environnement
Du 16/09/2024 de 11:00 à 12:15
R1-09
VAN BIESEBROECK Johannes (KU Leuven)
*Imperfect (Re)allocation in Imperfect Markets: Evidence from China’s Pilot Carbon ETS*
We evaluate the effectiveness of both absolute and intensity emission regulations in the Chinese carbon emission trading scheme (ETS) pilots that were established in 2013. The coexistence of two types of regulation within a single energy market provides a unique opportunity to study their differential effects on the magnitude and nature of emission reductions. We employ a difference-in-differences estimation strategy to examine the behavior of various market participants on all potential margins of adjustment, including energy consumption, industrial output, electricity trade, and substitution between power sources. We find that both types of regulation induce carbon mitigation, but that they come with distinct tradeoffs. Our results indicate that an aggregate annual reduction equivalent to an energy consumption of 7.5 and 8.7 million tons of standard coal can be attributed to the absolute and intensity-type emission regulations, respectively. This amounts to 9% and 23% of yearly energy consumption covered by the schemes.