Modelling the long-and-short run dynamics of share price movements in the resource sector

[page 49 missing] This paper involves the development and estimation of various statistical models that possess significant explanatory power in predicting future returns of resource shares. Models were constructed for IMPLATS, HARMONY, GOLDFIELDS, SASOL and SAPP!. Significant long-term relationships were found between company share price, earnings per share, dividends per share, commodity price and interest rates. The technique of co-integration permits these non-stationary economic variables to be linked by a stable long-term equilibrium relationship. For each co-integrating relationship an error correction model is developed to correct for short-term disequilibrium. The relationship between co-integration and error correction models is further extended to forecast future returns of share prices. A one-month rolling window forecasting method was implemented to forecast future returns of shares over one, two and three months respectively. Findings suggest that these results could be of considerable significance when making bets on a diversified portfolio of investments.

Grantor

School of Economics

Year dc:date.issued

2005

Author dc:creator

• Davids, Megan

Advisor dc:contributor.advisor

• van Rensburg, Paul

• Economics

Handle dc:identifier.uri

http://hdl.handle.net/11427/40149

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/40149