Modeling stock returns in the South African stock exchange: A nonlinear approach

Lumengo Bonga-Bonga, Michael Makakabule

Research output: Contribution to journalArticlepeer-review

11 Citations (Scopus)


This paper investigates the relationship between stock returns and macroeconomic variables, taking into account asymmetric adjustment behaviour in the stock market. The study applies the Smooth Transition Regression (STR) model to account for smooth asymmetric response of stock returns from economic variables. The results show that changes in dividend yield is an important factor in determining the asymmetric behaviour of stock returns on the South African stock market. Furthermore, the forecast performance of the STR model is compared with Ordinary Least Square (OLS) and Random Walk models. The STR, as a nonlinear model, outperforms the OLS and Random Walk models in an out-of-sample forecast. The findings of the paper violate the weak and semi-strong form test of the efficient market hypothesis.

Original languageEnglish
Pages (from-to)168-177
Number of pages10
JournalEuropean Journal of Economics, Finance and Administrative Sciences
Issue number19
Publication statusPublished - Apr 2010


  • Forecast
  • Smooth transition regression
  • Stock returns

ASJC Scopus subject areas

  • General Business,Management and Accounting
  • Economics, Econometrics and Finance (all)


Dive into the research topics of 'Modeling stock returns in the South African stock exchange: A nonlinear approach'. Together they form a unique fingerprint.

Cite this