This preview shows page 1. Sign up to view the full content.
Unformatted text preview: survey answers & vector 3 For each voter i …nd the nearest neighbor match j in the treatment
group H = h based on Mahalanobis distance on covariates 4 Take post-prior di¤erence in marginals for j . Apply the di¤erences to
i ’ priors to …nd the simulated posterior of i
s 5 Compute i ’ expected utilities and vote choice
Kendall, Nannicini & Trebbi (2014): “How Do Voters Respond to Information?” Motivation Empirical model Experimental design Reduced-form results Model estimation Counterfactual electoral campaigns
treatment only Predicted vote di¤erence
(in percentage points)
[0.77, 3.33] Blanket ideology
treatment only -2.2
[-3.37, -0.27] Blanket valence
plus ideology treatment 0.5
[-0.73, 1.84] Valence treatment to center & right
valence & ideology to left 1.3
[-0.19, 2.37] Ideology to center & right
valence & ideology to left -2.4
[-3.87, -0.92] Actual electoral
campaign e¤ect 1.8
[1.23, 3.14] Bootst...
View Full Document
This document was uploaded on 02/26/2014 for the course ECON 544 at The University of British Columbia.
- Winter '14