A thesis submitted in partial fulfilment of the requirements for the degree of
MASTER OF MANAGEMENT IN FINANCE AND INVESTMENTS
Of
WITS BUSINESS SCHOOL
March 2015 / The paper investigates the presence of two calendar anomalies; the day of the week or Mon-day effect and the Month of the year or January effect by modelling volatility of the industrial index returns on the Zimbabwe Stock Exchange (ZSE) pre and post the multi-currency sys-tem. The procedure is carried out by employing non-parametric models from the Generalized Autoregressive Conditional Heteroscedastic (GARCH) family; GARCH, Exponential GARCH (EGARCH) and Threshold GARCH (TGARCH). The models are better suited in modelling daily and monthly seasonality as they can capture the time-varying volatility of the stock return data. The period of analysis is from the January 2004 to April 2008 (pre-dollarization period) and the second period of analysis is from the post-currency reform which runs from February 2009 to December 2013.
The results obtained from the study are mixed. The day of the week test finds significantly negative returns on Monday, Wednesday and Friday pre the currency reform whilst a nega-tive Wednesday effect is found post the currency reform period. The TGARCH model is the only one that captures a negative monthly effects on all the months of the year with the ex-ception of January pre the currency reform period. No monthly effects are found on the ZSE post the currency reform period by all models employed. The absence of monthly seasonality effects and the reduced number of days of day of the week effects from all the GARCH mod-els employed can infer that the currency reform had a positive impact which translated to market efficiency.
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/20815 |
Date | 04 August 2016 |
Creators | Paradza, Abba |
Source Sets | South African National ETD Portal |
Language | English |
Detected Language | English |
Type | Thesis |
Format | application/pdf |
Page generated in 0.0073 seconds