Policy analysis with beliefs about identifying restrictions

Informazioni relative al progetto

PABIR

ID dell’accordo di sovvenzione: 682877

DOI

10.3030/682877

Progetto chiuso

Data della firma CE 22 Agosto 2016

Data di avvio 1 Ottobre 2016

Data di completamento 30 Settembre 2022

Finanziato da

• EXCELLENT SCIENCE - European Research Council (ERC)

Costo totale

€ 1 090 686,00

Contributo UE

€ 1 090 686,00

Coordinato da

UNIVERSITY COLLEGE LONDON
United Kingdom

Periodic Reporting for period 4 - PABIR (Policy analysis with beliefs about identifying restrictions)

Periodo di rendicontazione: 2021-04-01 al 2022-09-30

Sintesi del contesto e degli obiettivi generali del progetto

This project developed a toolkit of new econometric methods for policy analysis using Structural Vector Autoregressions (one of the leading approaches for conducting empirical macroeconomic policy analysis). The methods allow researchers to relax the strong
- and often dubious - assumptions that were previously made in the literature or to formally incorporate beliefs about the validity of such assumptions. The methods rely on Robust Bayesian analysis, which offers computational advantages and a formal way to incorporate beliefs, while removing the sensitivity to assumptions that plagued existing approaches.

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

The project resulted in several articles published in Econometrica, Journal of Econometrics, Quantitative Economics and Journal of Business and Economic Statistics, which developed the methods and illustrated their empirical relevance.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

One practical implication of the new methods is the suggestion to report set estimators, instead of point estimators. These sets correctly represent what can be learned from the data. The empirical findings in the published articles show that policy conclusions established in the literature can in fact be driven by arbitrary assumptions, thus offering a word of caution to empirical researchers as well as a strong motivation for using the new methods.

image.png