Regression discontinuity (RD) designs are usually interpreted as local randomized experiments. We formalize this concept proposing a probabilistic formulation of the assignment mechanism. We develop a Bayesian approach to draw inferences for the causal effects around the threshold, focussing on a fuzzy RD design where the receipt of the treatment is based on eligibility criteria and a voluntary application status. The method is applied to evaluate the effects of Italian university grants on student dropout.

Bayesian Inference for Fuzzy Regression Discontinuity Designs / Li F.; Mattei A.; Mealli F.. - ELETTRONICO. - (2013), pp. 1-8. (Intervento presentato al convegno SIS 2013 Statistical Conference: Advances in Latent Variables - Methods, Models and Applications tenutosi a Brescia nel 19-21 Giugno 2013).

Bayesian Inference for Fuzzy Regression Discontinuity Designs

MATTEI, ALESSANDRA;MEALLI, FABRIZIA
2013

Abstract

Regression discontinuity (RD) designs are usually interpreted as local randomized experiments. We formalize this concept proposing a probabilistic formulation of the assignment mechanism. We develop a Bayesian approach to draw inferences for the causal effects around the threshold, focussing on a fuzzy RD design where the receipt of the treatment is based on eligibility criteria and a voluntary application status. The method is applied to evaluate the effects of Italian university grants on student dropout.
2013
Advances in Latent Variables
SIS 2013 Statistical Conference: Advances in Latent Variables - Methods, Models and Applications
Brescia
19-21 Giugno 2013
Li F.; Mattei A.; Mealli F.
File in questo prodotto:
File Dimensione Formato  
LMM_SIS2013.pdf

Accesso chiuso

Descrizione: Articolo principale
Tipologia: Pdf editoriale (Version of record)
Licenza: Tutti i diritti riservati
Dimensione 134.53 kB
Formato Adobe PDF
134.53 kB Adobe PDF   Richiedi una copia

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/809084
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact