Day-of-the-week effect in Nigerian stock exchange: adaptive market hypothesis approach

  • Received November 21, 2019;
    Accepted December 18, 2019;
    Published March 2, 2020
  • Author(s)
  • DOI
  • Article Info
    Volume 17 2020, Issue #1, pp. 97-108
  • Cited by
    2 articles

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

The problems that this study informed are rooted in the uncertainty surrounding the presence of calendar anomalies in the Nigerian stock market and the need to ascertain whether calendar anomaly is changing with time and market condition according to the adaptive market hypothesis. This study evaluates how calendar anomaly behaves over time in the Nigerian stock market through the day-of-the-week effect since the latest trend is to examine time-changing anomaly. The general All Share Index returns of the Nigerian Stock Exchange between 2000 and 2017 are used in the analysis. Secondary daily index returns data for the period are sourced from the NSE Fact Book. The major estimation techniques employed in the study are the mean equations of the generalized autoregressive conditional heteroscedasticity (GARCH) and overlapping sub-period methodology. Moreover, returns are grouped into Up and Down periods depending on the periods that generate positive and negative returns, respectively. This study found out that Monday (MON), Tuesday (TUE), and Friday (FRI) effects are the only adaptive day-of-the-week effects. Thus, three (MON, TUE, FRI) day of the week effects found in the full sample are time-varying in subsample and are affected by market condition. On the whole, MON and Thursday (THUR) effects are found in Bull, while TUE and FRI are found in Bear. The investor must be careful to take time-variation into consideration; otherwise, they may incur a loss by thinking that the day-of-the-week effect is present every time.

view full abstract hide full abstract
    • Figure 1. Time-varying day-of-the-week effect
    • Table 1. Day-of-the-week effect results
    • Table 2. Results of time-varying day-of-the-week effect
    • Table 3. Results of day-of-the-week in Up and Down condition