• DocumentCode
    183069
  • Title

    Investor sentiment affects the trading volume: An evidence from network forums text analysis

  • Author

    Juanjuan Lai ; Tianlun Zheng ; Hongbo Yi ; Dayong Dong

  • Author_Institution
    Sch. of Econ. & Manage., Southwest Jiaotong Univ., Chengdu, China
  • fYear
    2014
  • fDate
    19-21 Aug. 2014
  • Firstpage
    666
  • Lastpage
    670
  • Abstract
    By conducting a word frequency statistics in the posing on Eastmoney forum bar about SSE Composite Index from January 7, 2010 to August 30, 2013, we establish a set of keyword dictionary to measure investor sentiment effectively, and accordingly to study the mutual relations between the abnormal investor sentiment based on the network forums and the abnormal trading volume in the trading market. The empirical results show that investor sentiment of the trading day is positive correlated with abnormal trading volume of the trading day. The abnormal investor sentiment of the previous trading day is negative correlated with abnormal trading volume of the next trading day. We also take the weekly data into consideration and find that there is no significant correlation between abnormal investor sentiment and market abnormal trading volume during the same period, while investor sentiment of the previous week and abnormal trading volume of the next week are negatively correlated.
  • Keywords
    investment; securities trading; statistics; text analysis; word processing; Eastmoney forum bar; SSE composite index; investor sentiment; market abnormal trading volume; network forum text analysis; trading market; word frequency statistics; Correlation; Dictionaries; Frequency measurement; Indexes; Mathematical model; Reactive power; Stock markets; abnormal trading volume; investor sentiment; word frequency statistics;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Fuzzy Systems and Knowledge Discovery (FSKD), 2014 11th International Conference on
  • Conference_Location
    Xiamen
  • Print_ISBN
    978-1-4799-5147-5
  • Type

    conf

  • DOI
    10.1109/FSKD.2014.6980914
  • Filename
    6980914