University of Sussex
Browse
Accepted version 20_09_21.pdf (1.32 MB)

Investor sentiment and the dispersion of stock returns: evidence based on the social network of investors

Download (1.32 MB)
journal contribution
posted on 2023-06-10, 01:02 authored by Alya Al-Nasseri, Faek Menla AliFaek Menla Ali, Allan Tucker
This paper extracts an investor sentiment indicator for the 30 DJIA stocks based on the textual classification of 289,024 online tweets posted on the so-called StockTwits, and examines its contemporaneous and predictability effects on the dispersion of stock returns using the quantile regression technique. We find that both contemporaneous and predictability effects of sentiment are heterogeneous throughout the return distribution. Specifically, sentiment is positively contemporaneously associated with stock returns at higher quantiles. However, it is a strong negative predictor of future returns at lower quantiles. Overall, our findings are broadly consistent with most behavioural theories and show that sentiment mainly affects the valuation of assets in extreme market conditions.

History

Publication status

  • Published

File Version

  • Accepted version

Journal

International Review of Financial Analysis

ISSN

1057-5219

Publisher

Elsevier

Volume

78

Article number

a101910

Department affiliated with

  • Accounting and Finance Publications

Full text available

  • Yes

Peer reviewed?

  • Yes

Legacy Posted Date

2021-09-21

First Open Access (FOA) Date

2023-03-26

First Compliant Deposit (FCD) Date

2021-09-20

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC