Evaluation of traditional and residual momentum strategies during the Covid period on the Johannesburg Stock Exchange
- Authors: Yengwa, Mphathi Lubabalo
- Date: 2024-10-11
- Subjects: Efficient market theory , Residual momentum , Economic crisis , Johannesburg Stock Exchange , COVID-19 Pandemic, 2020-2023 Influence
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/462834 , vital:76339
- Description: Traditional momentum is a concept which was first discovered by Jegadeesh and Titman (1993), defined as a tendency of stocks to experience a continuation in their relative performance. A stock that performed relatively well will continue to perform relatively well, and vice versa. It has been observed by other researchers that during market crises, traditional momentum tends to produce large negative returns for investors, defined as a momentum crash. To mitigate momentum crashes, many researchers have developed new momentum strategies which have better performance than traditional momentum during market crises; such strategies include residual momentum. While both residual and traditional momentum have been studied in international markets and locally, the performance of both the residual and traditional momentum strategies have not been examined in the most recent Covid-fuelled financial crisis on the Johannesburg Stock Exchange. The study compares the performance of hypothetical long-only winner traditional and residual momentum portfolios (from 2018–2022) using various risk metrics, which include the tracking error, Sharpe ratio, Jensen’s alpha and information ratio. To compare the statistical significance of the difference in mean returns of residual and traditional momentum strategies to the benchmark (FTSE/Johannesburg Stock Exchange (JSE) Top 40) the study uses Welch’s t-test. The study uses an Auto regressive distributed lag (ARDL) regression to examine the effect that various market conditions (bull market, bear market and extreme volatility) have on the returns of residual and traditional momentum strategies. Given the limited period examined in this study, the Monte Carlo simulation was used to extrapolate potential outcomes of how the momentum strategies might perform under different market conditions (as mentioned) in 1 000 iterations of each condition. The simple return analysis undertaken in this research revealed that traditional momentum outperformed residual momentum both before and throughout the COVID period. In the risk-adjusted performance measures, traditional momentum outperformed at all four risk indicators during the 2020 COVID year. The statistical significance tests, which compared the strategies' mean returns to the benchmark, demonstrated no statistically significant difference in returns over the COVID year. Furthermore, when evaluating the strategies over a five-year period (2018-2022), the difference in mean returns was shown to be statistically insignificant. However, statistical significance in returns was shown in some individual years. The ARDL regression findings show that bull, bear, and volatility factors explain relatively little of the returns for both momentum strategies, which is consistent with previous research. The Monte Carlo simulation, using the bear variable, forecasted that traditional momentum would result in negative returns during market declines, but residual momentum would provide positive returns and surpass traditional momentum with a probability of 26%. When using the bull variable, the simulation discovered that both traditional and residual momentum strategies resulted in positive returns. However, the residual momentum strategy outperformed in terms of returns and had an 84% likelihood of outperforming the traditional momentum strategy across 1,000 iterations. Nevertheless, when the simulation included the volatility variable, it projected negative returns for residual momentum and positive returns for traditional momentum. Additionally, it estimated a 14% probability of residual momentum surpassing traditional momentum under volatile market circumstances. , Thesis (MCom) -- Faculty of Commerce, Economics and Economic History, 2024
- Full Text:
- Date Issued: 2024-10-11
- Authors: Yengwa, Mphathi Lubabalo
- Date: 2024-10-11
- Subjects: Efficient market theory , Residual momentum , Economic crisis , Johannesburg Stock Exchange , COVID-19 Pandemic, 2020-2023 Influence
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/462834 , vital:76339
- Description: Traditional momentum is a concept which was first discovered by Jegadeesh and Titman (1993), defined as a tendency of stocks to experience a continuation in their relative performance. A stock that performed relatively well will continue to perform relatively well, and vice versa. It has been observed by other researchers that during market crises, traditional momentum tends to produce large negative returns for investors, defined as a momentum crash. To mitigate momentum crashes, many researchers have developed new momentum strategies which have better performance than traditional momentum during market crises; such strategies include residual momentum. While both residual and traditional momentum have been studied in international markets and locally, the performance of both the residual and traditional momentum strategies have not been examined in the most recent Covid-fuelled financial crisis on the Johannesburg Stock Exchange. The study compares the performance of hypothetical long-only winner traditional and residual momentum portfolios (from 2018–2022) using various risk metrics, which include the tracking error, Sharpe ratio, Jensen’s alpha and information ratio. To compare the statistical significance of the difference in mean returns of residual and traditional momentum strategies to the benchmark (FTSE/Johannesburg Stock Exchange (JSE) Top 40) the study uses Welch’s t-test. The study uses an Auto regressive distributed lag (ARDL) regression to examine the effect that various market conditions (bull market, bear market and extreme volatility) have on the returns of residual and traditional momentum strategies. Given the limited period examined in this study, the Monte Carlo simulation was used to extrapolate potential outcomes of how the momentum strategies might perform under different market conditions (as mentioned) in 1 000 iterations of each condition. The simple return analysis undertaken in this research revealed that traditional momentum outperformed residual momentum both before and throughout the COVID period. In the risk-adjusted performance measures, traditional momentum outperformed at all four risk indicators during the 2020 COVID year. The statistical significance tests, which compared the strategies' mean returns to the benchmark, demonstrated no statistically significant difference in returns over the COVID year. Furthermore, when evaluating the strategies over a five-year period (2018-2022), the difference in mean returns was shown to be statistically insignificant. However, statistical significance in returns was shown in some individual years. The ARDL regression findings show that bull, bear, and volatility factors explain relatively little of the returns for both momentum strategies, which is consistent with previous research. The Monte Carlo simulation, using the bear variable, forecasted that traditional momentum would result in negative returns during market declines, but residual momentum would provide positive returns and surpass traditional momentum with a probability of 26%. When using the bull variable, the simulation discovered that both traditional and residual momentum strategies resulted in positive returns. However, the residual momentum strategy outperformed in terms of returns and had an 84% likelihood of outperforming the traditional momentum strategy across 1,000 iterations. Nevertheless, when the simulation included the volatility variable, it projected negative returns for residual momentum and positive returns for traditional momentum. Additionally, it estimated a 14% probability of residual momentum surpassing traditional momentum under volatile market circumstances. , Thesis (MCom) -- Faculty of Commerce, Economics and Economic History, 2024
- Full Text:
- Date Issued: 2024-10-11
The stock market and the business cycle in South Africa
- Authors: Pokoo, Patience
- Date: 2024-10-11
- Subjects: Stock exchanges South Africa , Economic activity , Business cycles South Africa , Autoregression (Statistics) , Policymaker , Johannesburg Stock Exchange
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/462801 , vital:76336
- Description: The relationship between the stock market and economic activity has long been a topic for research. Several studies done in both advanced and emerging economies including South Africa before COVID-19 found stock market prices predict the cycle of real economic activity and some found it to be the reversal. Therefore, this Study seeks to examine this topic and will extend beyond the post-covid period exploring the relationship between the stock market (proxied by the JSE All-Share Index) and the business cycle (represented by the Coincident Business Cycle Indicator of the SARB) in South Africa. The study also investigates if the relationship between the stock market and the business cycle is homogenous across the three selected sectors of the JSE using a combination of the “financial accelerator theory”, the “wealth effect theory”, the “traditional valuation model of stock prices”, the “stock prices as aggregators of expectations”, and the “cost of raising equity capital”. The Econometrics models employed include time-series and panel cointegration techniques, relying on the ARDL estimation model and a Granger-Causality Test. The findings of this study indicate that a long-run relationship exists between the stock market and the business cycle in South Africa. The findings support the notion that the stock market predicts economic activity, and this relationship is assumed to be homogenous across the selected Sectors of the JSE (namely, Resources, Financials, and Industrials). Again, the Granger-Causality Test confirms the relationship between the stock market and the business cycle in South Africa to be unidirectional. It is recommended that since the stock market affects South African economic activity positively in the long run which is consistent with findings of similar studies done on the JSE, the South African Reserve Bank (SARB) must strengthen existing policy to ensure financial system stability and sustainable economic growth in South Africa. Again, the stock market being a leading indicator of the business cycle is something different. As a recommendation, we need to look at ways to use the prediction ability in a business setting. Investors and Portfolio Managers can follow trends of the stock market to forecast the direction of the future economy to make educated decisions to hedge their investments and diversify their portfolios against huge losses in crises such as the Financial Crises and the Global Health Crisis (COVID-19), however, with the caveat that the stock market does not always accurately predict the business cycle. , Thesis (MCom) -- Faculty of Commerce, Economics and Economic History, 2024
- Full Text:
- Date Issued: 2024-10-11
- Authors: Pokoo, Patience
- Date: 2024-10-11
- Subjects: Stock exchanges South Africa , Economic activity , Business cycles South Africa , Autoregression (Statistics) , Policymaker , Johannesburg Stock Exchange
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/462801 , vital:76336
- Description: The relationship between the stock market and economic activity has long been a topic for research. Several studies done in both advanced and emerging economies including South Africa before COVID-19 found stock market prices predict the cycle of real economic activity and some found it to be the reversal. Therefore, this Study seeks to examine this topic and will extend beyond the post-covid period exploring the relationship between the stock market (proxied by the JSE All-Share Index) and the business cycle (represented by the Coincident Business Cycle Indicator of the SARB) in South Africa. The study also investigates if the relationship between the stock market and the business cycle is homogenous across the three selected sectors of the JSE using a combination of the “financial accelerator theory”, the “wealth effect theory”, the “traditional valuation model of stock prices”, the “stock prices as aggregators of expectations”, and the “cost of raising equity capital”. The Econometrics models employed include time-series and panel cointegration techniques, relying on the ARDL estimation model and a Granger-Causality Test. The findings of this study indicate that a long-run relationship exists between the stock market and the business cycle in South Africa. The findings support the notion that the stock market predicts economic activity, and this relationship is assumed to be homogenous across the selected Sectors of the JSE (namely, Resources, Financials, and Industrials). Again, the Granger-Causality Test confirms the relationship between the stock market and the business cycle in South Africa to be unidirectional. It is recommended that since the stock market affects South African economic activity positively in the long run which is consistent with findings of similar studies done on the JSE, the South African Reserve Bank (SARB) must strengthen existing policy to ensure financial system stability and sustainable economic growth in South Africa. Again, the stock market being a leading indicator of the business cycle is something different. As a recommendation, we need to look at ways to use the prediction ability in a business setting. Investors and Portfolio Managers can follow trends of the stock market to forecast the direction of the future economy to make educated decisions to hedge their investments and diversify their portfolios against huge losses in crises such as the Financial Crises and the Global Health Crisis (COVID-19), however, with the caveat that the stock market does not always accurately predict the business cycle. , Thesis (MCom) -- Faculty of Commerce, Economics and Economic History, 2024
- Full Text:
- Date Issued: 2024-10-11
The yield spread as a predictor for buy or sell signals for sectoral indices of the JSE
- Authors: Roeber, Christine
- Date: 2023-10-13
- Subjects: Yield curve , Rate of return South Africa , Yield spread , Interest rate , Johannesburg Stock Exchange
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/419687 , vital:71666
- Description: The predictive nature of the yield curve has been of interest to researchers for years. In this thesis, the evidence for the yield curve as a predictor is examine, specifically as a predictor for bear markets in the JSE stock market for 8 sub-sectoral indices. The study explores a dynamic market timing strategy for timing the South African stock market compared to a normal buy-and-hold strategy. First, probit models are estimated for each of the sectoral indices which did not prove to have tracked well all the bear market phases. Then a dynamic market timing portfolio is simulated against a buy-and-hold only strategy, the dynamic market timing portfolio proved to have outperformed a buy-and-hold strategy for almost all the indices. Thus, a Henriksson-Merton parametric model test which tests for market timing ability was done on these sub-indices. The research finds that the yield curve in South Africa is not a useful tool for a buy-sell strategy for most of the sub-sectoral indices of the JSE. , Thesis (MCom) -- Faculty of Commerce, Economics and Economic History, 2023
- Full Text:
- Date Issued: 2023-10-13
- Authors: Roeber, Christine
- Date: 2023-10-13
- Subjects: Yield curve , Rate of return South Africa , Yield spread , Interest rate , Johannesburg Stock Exchange
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/419687 , vital:71666
- Description: The predictive nature of the yield curve has been of interest to researchers for years. In this thesis, the evidence for the yield curve as a predictor is examine, specifically as a predictor for bear markets in the JSE stock market for 8 sub-sectoral indices. The study explores a dynamic market timing strategy for timing the South African stock market compared to a normal buy-and-hold strategy. First, probit models are estimated for each of the sectoral indices which did not prove to have tracked well all the bear market phases. Then a dynamic market timing portfolio is simulated against a buy-and-hold only strategy, the dynamic market timing portfolio proved to have outperformed a buy-and-hold strategy for almost all the indices. Thus, a Henriksson-Merton parametric model test which tests for market timing ability was done on these sub-indices. The research finds that the yield curve in South Africa is not a useful tool for a buy-sell strategy for most of the sub-sectoral indices of the JSE. , Thesis (MCom) -- Faculty of Commerce, Economics and Economic History, 2023
- Full Text:
- Date Issued: 2023-10-13
- «
- ‹
- 1
- ›
- »