Skip to main navigation Skip to search Skip to main content

Seeking alpha from dividend announcements: Big-data insights from joining CAR and EVA style analysis

  • University of Massachusetts Boston
  • Northeastern University

Research output: Contribution to journal › Article › peer-review

Abstract

The authors conduct a big-data analysis to assess the pricing impact of dividend announcements. They join the traditional cumulative abnormal return (CAR) approach to estimating abnormal returns with economic value added (EVA) style analysis. The empirical results are significant for a sample of 172,481 dividend announcements during the 23 years from 1997–2019. For active investors, alpha-generating trading returns may be available by longing the stocks of dividend-increasing companies and shorting the stocks of dividend-deceasing companies. A notable finding linked to EVA style analysis is that alpha returns on dividend-increasing companies were available on the stocks of value-creating growth companies, the stocks of underinvesting companies, and especially the stocks of “wise contraction or cyclical restructuring” companies in need of a turnaround growth signal. Alpha returns on dividend-decreasing companies were available on the stocks of value-destroying growth companies and the “wise restructuring” companies. The authors also find significant dividend announcement CARs and potential alpha-generating trading opportunities across 10 S&P-classified sectors.

Original languageEnglish
Pages (from-to)1-20
Number of pages20
JournalJournal of Investing
Volume30
Issue number5
DOIs
StatePublished - Aug 2021

ASJC Scopus Subject Areas

  • Finance
  • Strategy and Management
  • Management of Technology and Innovation

Keywords

  • Big data/machine learning performance measurement*
  • Quantitative methods
  • Security analysis and valuation

Fingerprint

Dive into the research topics of 'Seeking alpha from dividend announcements: Big-data insights from joining CAR and EVA style analysis'. Together they form a unique fingerprint.

Cite this