African Journals Online
Studies in Economics and Econometrics

Volume 26, Issue 3, November 2002
ABSTRACTS

The South African share index futures and share markets: efficiency and causality revisited

Leng, H.M.J.

Abstract: In this paper the efficiency of the stock index futures market and the underlying spot market is investigated over the period January 2, 1996 to June 29, 2001. In addition to examining the whole data set, the sample was also divided into four subsamples, each of which was analysed independently. The four subsamples included a pre-crisis or tranquil period, a crisis period, a transition period and, finally, a post-crisis period. The crisis period spans the time of the Asian financial crisis whereas the transition period incorporates two events which may have impacted on market efficiency - the Russian ruble crisis and the South African rand crisis. We tested for unit roots, market cointegration and Granger causality with vector error correction and found that the two markets were relatively inefficient during the pre-crisis period but that efficiency improved following the onset of the crisis. The futures market proved to be the dominant market in the long-run during all four subsample periods since it was primarily responsible for maintaining its equilibrium relationship with the underlying spot market. Over the short-term, the futures return was found to lead or Granger-cause the spot return during the pre-crisis period, but this was reversed during the crisis period. No short-term causality relationship could be detected for the transition and post-crisis periods. Futures prices are expected to lead spot prices because of lower transaction costs and less restrictive short-selling in the futures market. We speculate that the anomaly, i.e., the spot price lead, during the crisis period reflect investor behaviour during times of uncertainty and distress when more value is attached to the underlying asset than to past performance or future prospects of the derivative instrument.

South African unit trusts: selection ability and information effects

Oosthuizen, H.R.; Smit, E. vd M.

Abstract: Research into unit trusts are often concerned with the ability of fund managers to achieve superior performance. However, the growing popularity of unit trusts as an investment vehicle may also lead one to question the ability of individual unit trust investors themselves. The main research objective of this study is to establish whether South African unit trust investors display selection ability, i.e. whether investors are smart ex ante, in that they move to funds that will perform better. The secondary research objective is to establish whether investors' moves (in the form of cash flows into and out of funds) contain information that can be utilised to earn abnormal returns - the so-called information effect. The analysis is based on a performance test introduced by Grinblatt and Titman (1993) and further developed and applied by Zheng (1999). Evidence is presented that confirm that investors display selection ability, although this ability is very weak. There is, however, no significant evidence to support the information effect. Tests conducted to examine the information effect provide some evidence that money flows from funds that subsequently underperform the market.

The term structure as a predictor of recessions

Moolman, E.

Abstract: Despite the existence of macroeconomic models and complex business cycle indicators, policymakers and market participants can benefit by looking at a few well-chosen indicators such as the term structure of interest rates in predicting business cycle turning points. If the term structure accurately predicts business cycle turning points, it provides an easy way to confirm the predictions of macroeconomic models, or it can eliminate the need for a macroeconomic model the interest is in the turning points and not in the levels of the business cycle. The objective of this paper is to predict turning points of the South African business cycle based on its relationship with the term structure of interest rates. A probit model was used, and the results indicate that the term structure successfully predicts turning points of business cycle two quarters ahead. The negative empirical relationship between the term structure of interest rates and the business cycle conforms to economic theory.

Using demographic and health surveys to measure poverty - an application to South Africa

Booysen, F. le R.

Abstract: There are different approaches to the measurement of poverty. These depend on the objective of the analysis, the nature of the data and the method employed in measuring poverty. The asset index approach applied to data from Demographic and Health Surveys (DHS) has gained increasing popularity in recent years, particularly in analyses of the relationship between poverty, health and population issues. The results presented in this paper suggest that it is possible, in the absence of income and expenditure data, to employ the available data from the DHS to measure differences in the socioeconomic status of South African households. The asset index represents an internally coherent, robust and comparable indicator of poverty. An analysis of the relationship between poverty, health and population issues, for which the DHS data set presents a wealth of data, will be impossible without such an asset index.

An application of self-organising and backpropagation neural networks for predicting segment classification

Bloom, J.Z.

Abstract: Inadequate market segmentation and clustering problems could cause an enterprise to either miss a strategic marketing opportunity or not cash-in on a tactical campaign. The need for in-depth knowledge of customer segments and the need to overcome the limitations of using linear techniques to analyse non-linear problems requires a reassessment of generally used approaches. The objectives of the research are (1) to consider the use of self-organising (SOM) neural networks for segmenting customer markets and (2) to analyse the predictive ability of backpropagation (BP) neural networks for classifying new customers by using the output provided by SOM neural networks. The nature and scope of neural networks are considered and the conceptual differences between Cluster Analysis and SOM neural networks as well as BP neural networks and multiple linear regression (MLR) static filter model are highlighted. The findings of the SOM neural network modelling indicate four natural clusters. In addition, the predictive ability of the BP neural network model was superior to that of MLR. Additional knowledge was also extracted from the BP neural network model by analysing the relationship between the input variables and each segment by means of input-output analysis. Sensitivity analysis was also used to identify important variables for each segment. Input-output analysis is also used to compile broad profiles of differences between the segments. The BP neural network model developed for this application is also suitable for deployment (i.e. classification of "new" customers) with a high level of confidence.