An Assessment of the Application of Various Models in Share Selection, Portfolio Construction and Analysis among East African Stock Exchanges
by Shamis Said Moh’d
The purpose of this study is to examine the applicability of equity portfolios in various East African stock exchanges using Markowitz portfolio model and Capital Asset Pricing model (CAPM), that can be used by institutional investors and young people within the region in making investment decisions, broaden their knowledge of quantitative techniques of portfolio construction and narrow the heuristic commonsense approach.
The value investing theory, modern portfolio theory, capital market theory and Tobin separation theory are the underpinned theories used in this study. The related models of these theories such Mean Variance Covariance Model (MVCM) and Capital Assets Pricing Model (CAPM) were separately merged with Data Envelopment Analysis (DEA) and the resulted hybrid models which are DEAMVCM and DEA-CAPM was used in this study.
The efficient assessment of the questions in this study can be attained using a quantitative research design and more specifically descriptive design. The data related to country economy, economic sectors, capital markets, companies’ fundamentals, closing share prices, markets indices and exchange rates from 2015 to 2018 were collected in respective database available online. A judgemental sampling technique was adapted while collecting data, whereby, out of 115 listed companies only 52 are found to have all data required in specified time frame.
The MATLAB program was used to analyse the data. Furthermore, the independent sample t test and ANOVA were used to measure the statistical significance of the hypothesis. In relation to share selection, the findings revealed that all listed companies from USE, RSE, and all companies from the service and manufacturing sectors in all capital markets were not attained the minimum performance stated, therefore were excluded for further analysis. Only some companies that fall under the industry sector in Kenya and Tanzania were shortlisted as their combined performance is equal or above the benchmark. Out of 52 companies, only 11 companies were qualified for further analysis. Among them, 6 companies are from Kenya which is equivalent to 16 percent of the total companies, and 5 from Tanzania which is equivalent to 56 percent of the total companies evaluated from Kenya and Tanzania, respectively. Impliedly, combining the performance of country economy, economic sectors and company’s fundamentals has a major impact on screening the stocks to be used for portfolio construction.
The results further revealed that portfolio construction, performance and optimization varied with portfolio size and model used. Moreover, different states of economy generated different portfolio risk and returns. Therefore, it is recommended to the management of capital markets, regulatory bodies and listed companies to ensure managerial and operational performance of the institutions. Since this study was limited to short time frame due to data availability future studies can incorporate more data range. Likewise, this study employs a bottom-up approach to combine various components such as country economy, stock markets development, economic sectors, and company fundamentals instead of using both bottom-up and top-down approaches. Although the methodology is still new in the field of stock selection, also limited literature demonstrated quantitatively performance evaluation of each component yet is recommended by the scholars when the combined components
involve more than one country.