The causal linkages between investor sentiment and excess returns on Borsa Istanbul

ÇAĞLI E. Ç., ERGÜN Z. C., DURUKAN SALI M. B.

BORSA ISTANBUL REVIEW, vol.20, no.3, pp.214-223, 2020 (SSCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 20 Issue: 3
  • Publication Date: 2020
  • Doi Number: 10.1016/j.bir.2020.02.001
  • Journal Name: BORSA ISTANBUL REVIEW
  • Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, EconLit, Directory of Open Access Journals
  • Page Numbers: pp.214-223
  • Keywords: Behavioral finance, Investor sentiment, Granger causality, Recursive evolving algorithm, CHINESE STOCK-MARKET, CONSUMER CONFIDENCE, GRANGER CAUSALITY, IMPACT, MONEY
  • Dokuz Eylül University Affiliated: Yes

Abstract

The main aim of this study is to analyze the causal relationship between BIST-100 return index and investor sentiment. The investor sentiment is measured by constructing an index comprised of the closed-end fund discount, mutual fund flows, share of equity issues in aggregate issues, repo shares in mutual funds and turnover ratio on a monthly basis for the period from 1997 to 2018. We employ a novel Granger causality test developed by Shi, Phillips, and Hum (2018) which detects and dates the changes in causal relationships. The results show that the conventional Granger causality test indicates no causality between the sentiment index and BIST-100 return index. However, the recursive evolving window procedure detects Granger causality episodes for the proxies, except repo shares in mutual funds. The findings indicate that considering nonlinearities for the sample period could change the causal relationship between investor sentiment and the market return. Copyright (C) 2020, Borsa Istanbul Anonim Sirketi. Production and hosting by Elsevier B.V.