Mitigating misleading implications for policy: Treatment of outliers in a difference-indifferences framework

Alessio Reghezza (), Jonathan Williams and Philip Molyneux ()

No 19014, Working Papers from Bangor Business School, Prifysgol Bangor University (Cymru / Wales)

Abstract: Applications of the difference-in-differences estimator in economics, banking and finance, and management commonly treat outliers using the winsorize method. However, failure to winsorize outliers in both the treatment and controls groups introduces volatility in estimated coefficients, significance levels, and standard errors. A faulty process can lead to an exogenous event realising a significant effect that proper process would fail to detect. In demonstration, we randomly generate placebo interventions in bank-level data and discuss how to detect and limit the problem.

Keywords: Winsor; Difference-in-Differences; Outliers; Financial Data; Research Design (search for similar items in EconPapers)
JEL-codes: C00 C10 C13 C18 C19 C80 (search for similar items in EconPapers)
Date: 2019-05
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.bangor.ac.uk/business/research/documents/BBSWP-19-14.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bng:wpaper:19014

Access Statistics for this paper

More papers in Working Papers from Bangor Business School, Prifysgol Bangor University (Cymru / Wales) Contact information at EDIRC.
Bibliographic data for series maintained by Alan Thomas ( this e-mail address is bad, please contact ).