Is a sentiment-based trading strategy profitable?

ABSTRACT

We examine whether sentiment indices predict individual firms’ stock returns and evaluate the performances of sentiment-based trading strategies in the Korean equity market. We find that the sentiment indices (constructed using the principal component analysis (PCA) and overnight stock returns) positively predict stock price movements, whereas news sentiment does not significantly determine future stock returns. A comparison of portfolio performances among sentiment indices reveals that the long-short equity strategy based on PCA sentiment changes yields the highest return – a result that is not explained by well-known risk factors. Moreover, investors may earn even greater returns by employing multiple sentiment measures when constructing portfolios, suggesting that each measure reflects different aspects of investor sentiment.

KEYWORDS:

• cross-sectional stock returns
• investor sentiment
• portfolio management
• return predictability
• trading strategy

JEL CLASSFICATION:

• G10
• G11
• G40

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 For further information regarding the construction of the NSI, refer to the Economic Statistics System of the BOK (https://ecos.bok.or.kr/).

2 Sources: Dataguide (https://www.dataguide.co.kr/DG5web/eng/index.asp) and Handa Partners (https://sites.google.com/view/handaresearch/handa-partners).