Browsing by Author "Huang, Chun-Sung"

Now showing 1 - 11 of 11
  • Thumbnail Image
    Item
    Open Access
    An Analysis of the Low-Volatility Anomaly on the Johannesburg Stock Exchange
    (2019) Harrisberg, Richard; Huang, Chun-Sung
    The low-volatility anomaly can be described as the unexpected outperformance of low-volatility stocks compared to high-volatility stocks over the long-term. This dissertation investigates the low-volatility anomaly and its presence on the Johannesburg Stock Exchange (JSE). Possible reasons behind why low-volatility stocks consistently outperform their high volatility counterparts, as well as their own expected return, over the long-term are discussed. This includes analysing how financial risk is measured and whether this plays a role in obscuring the expected risk-return relationship, in addition to other fundamental factors impacting expected returns. It is found that the low-volatility anomaly is present on the JSE and that using a number of different risk metrics does not significantly change where a stock is ranked on the risk spectrum. Additionally, including an interest rate exposure factor, a value factor and a momentum factor lowers the unexpected portion (Alpha) of the returns of low volatility stocks, at the same time as narrowing the gap between the unexpected performance of the lowest and highest volatility stocks.
  • Thumbnail Image
    Item
    Open Access
    Factors of the term structure of sovereign yield spreads and the effect on the uncovered interest rate parity model for exchange rate prediction
    (2018) Reddy, Desigan; Huang, Chun-Sung
    Using a Principal Component Analysis (PCA) approach, we investigate the sovereign yield spread term structure of the BRICS economies against the U.S. We show that the term structure for these markets are primarily driven by three latent factors which can be classified as the spread level, slope and curvature factors. We further postulate that a country’s yield curve contains valuable information about its future economic state and as such the PCA derived spread factors, which are based on the differences between sovereign yield curves, encapsulates material macro-economic information between the countries. In light of this, we show that augmenting the traditional Uncovered Interest Rate Parity model (UIRP) with these factors improves the models predictive accuracy of exchange rate movements.
  • Thumbnail Image
    Item
    Open Access
    Forecasting and modelling the VIX using Neural Networks
    (2022) Netshivhambe, Nomonde; Huang, Chun-Sung
    This study investigates the volatility forecasting ability of neural network models. In particular, we focus on the performance of Multi-layer Perceptron (MLP) and the Long Short Term (LSTM) Neural Networks in predicting the CBOE Volatility Index (VIX). The inputs into these models includes the VIX, GARCH(1,1) fitted values and various financial and macroeconomic explanatory variables, such as the S&P 500 returns and oil price. In addition, this study segments data into two sub-periods, namely a Calm and Crisis Period in the financial market. The segmentation of the periods caters for the changes in the predictive power of the aforementioned models, given the dierent market conditions. When forecasting the VIX, we show that the best performing model is found in the Calm Period. In addition, we show that the MLP has more predictive power than the LSTM.
  • Thumbnail Image
    Item
    Open Access
    Fourier pricing of two-asset options: a comparison of methods
    (2018) Roberts, Jessica Ellen; Ouwehand, Peter; Huang, Chun-Sung
    Fourier methods form an integral part in the universe of option pricing due to their speed, accuracy and diversity of use. Two types of methods that are extensively used are fast Fourier transform (FFT) methods and the Fourier-cosine series expansion (COS) method. Since its introduction the COS method has been seen to be more efficient in terms of rate of convergence than its FFT counterparts when pricing vanilla options; however limited comparison has been performed for more exotic options and under varying model assumptions. This paper will expand on this research by considering the efficiency of the two methods when applied to spread and worst-of rainbow options under two different models - namely the Black-Scholes model and the Variance Gamma model. In order to conduct this comparison, this paper considers each option under each model and determines the number of terms until the price estimate converges to a certain level of accuracy. Furthermore, it tests the robustness of the pricing methodologies to changes in certain discretionary parameters. It is found that although under the Black-Scholes model the COS method converges in fewer terms than the FFT method for both spread options (32 versus 128 terms) and the rainbow options (64 versus 512 terms), this is not the case under the more complex Variance Gamma model where the terms to convergence of both methods are similar. Both the methodologies are generally robust against changes in the discretionary variables; however, a notable issue appears under the implementation of the FFT methodology to worst-of rainbow options where the choice of the truncated integration region becomes highly influential on the ability of the method to price accurately. In sum, this paper finds that the improved speed of the COS method against the FFT method diminishes with a more complex model - although the extent of this can only be determined by testing for increasingly complex characteristic functions. Overall the COS method can be seen to be preferable from a practical point of view due to its higher level of robustness.
  • Thumbnail Image
    Item
    Open Access
    Highly efficient pricing of exotic derivatives under mean-reversion, jumps and stochastic volatility
    (2018) Huang, Chun-Sung; Mataramvura, Sure; O'Hara, John
    The pricing of exotic derivatives continues to attract much attention from academics and practitioners alike. Despite the overwhelming interest, the task of finding a robust methodology that could derive closed-form solutions for exotic derivatives remains a difficult challenge. In addition, the level of sophistication is greatly enhanced when options are priced in a more realistic framework. This includes, but not limited to, utilising jump-diffusion models with mean-reversion, stochastic volatility, and/or stochastic jump intensity. More pertinently, these inclusions allow the resulting asset price process to capture the various empirical features, such as heavy tails and asymmetry, commonly observed in financial data. However, under such a framework, the density function governing the underlying asset price process is generally not available. This leads to a breakdown of the classical risk-neutral option valuation method via the discounted expectation of the final payoff. Furthermore, when an analytical expression for the option pricing formula becomes available, the solution is often complex and in semi closed-form. Hence, a substantial amount of computational time is required to obtain the value of the option, which may not satisfy the efficiency demanded in practice. Such drawbacks may be remedied by utilising numerical integration techniques to price options more efficiently in the Fourier domain instead, since the associated characteristic functions are more readily available. This thesis is concerned primarily with the efficient and accurate pricing of exotic derivatives under the aforementioned framework. We address the research opportunity by exploring the valuation of exotic options with numerical integration techniques once the associated characteristic functions are developed. In particular, we advocate the use of the novel Fourier-cosine (COS) expansions, and the more recent Shannon wavelet inverse Fourier technique (SWIFT). Once the option prices are obtained, the efficiency of the two techniques are benchmarked against the widely-acclaimed fast Fourier transform (FFT) method. More importantly, we perform extensive numerical experiments and error analyses to show that, under our proposed framework, not only is the COS and SWIFT methods more efficient, but are also highly accurate with exponential rate of error convergence. Finally, we conduct a set of sensitivity analyses to evaluate the models’ consistency and robustness under different market conditions
  • No Thumbnail Available
    Item
    Open Access
    Machine Learning with FEARS index: does the inclusion of investor sentiment improve a machine learning model's ability to predict volatility?
    (2024) James, Andrew Michael; Huang, Chun-Sung
    The aim of this study is to determine whether the inclusion of investor sentiment allows machine learning methods to produce improved predictions of volatility in equity markets. Specifically, the investor sentiment measure is constructed as an index by using search volume data of different search terms obtained from Google Trends. The resulting Financial and Economic Attitudes Revealed by Search (FEARS) index is then utilised as a feature to forecast volatility via three different machine learning (ML) techniques, namely the Random Forest, Artificial Neural Network (ANN), and Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) methods. A consolidated dataset, where all G7 countries were combined into a single series, as well as an individualised dataset, where each individual country is analysed independently, were used to test the different ML methods' volatility forecasting ability. Our results show that, for the consolidated dataset, the inclusion of the FEARS index does not provide significant additional predictive power. However, through the individualised dataset, the FEARS index was shown in certain cases to provide greater predictive accuracy. Furthermore, it was observed that the LSTM-RNN outperformed the ANN and Random Forest methods, which indicates that our volatility prediction indeed benefits from elements of prior periods' volatilities as feature variables.
  • Thumbnail Image
    Item
    Open Access
    NPL forecasting under a fourier residual modified model: An empirical analysis of an unsecured consumer credit provider in South Africa
    (2016) Luckan, Pranisha; Huang, Chun-Sung
    Forecasting nonperforming loans (NPLs) is a primary objective for credit providers. NPL forecasts assist in financial budgeting and provisioning for bad debts. The difficulty in accurately identifying the determinants of domestic NPLs has led to a review of time series forecasting techniques. This dissertation explores whether a forecasting model combining a traditional time series approach with a Fourier series residual modification technique performs well in projecting NPLs. It also seeks to establish if selecting an adequate time series model before modifying its residual terms is of benefit. Using the data of an unsecured consumer credit provider in South Africa, the in-sample and out-of-sample performance for a seasonal time series model and residual modified model were evaluated. The results demonstrate that a time series model performs well but the out-of-sample forecasting errors may be reduced by including the lowest Fourier frequencies to modify the residual terms.
  • No Thumbnail Available
    Item
    Open Access
    Performance determinants of foreign currency bonds issued by JSE-listed companies? An indexation approach
    (2023) Neethling, Sean; Huang, Chun-Sung; Rajaratnam Kanshukan
    The number of Johannesburg Stock Exchange (JSE) listed companies that have issued foreign or hard currency denominated bonds has increased meaningfully since the global financial crisis in 2008. This broader hard currency opportunity set is primarily available to professional local investors but accessibility is largely constrained by a lack of reliable data and a representative index to assess performance. This thesis constructs hard-currency non-Rand denominated corporate bond indices using both traditional and fundamental indexation strategies to critically investigate the performance of the investable opportunity set. The constructed indices represent a first attempt to study the hard currency universe in South Africa and covers the 12-year period between 2008 and 2019. The main findings show that the constructed fundamental indices not only generate superior absolute nominal and excess returns versus a traditional market cap equivalent, but do so by taking less total, systematic and downside risk. The next substantive chapter considers the volatility of total bond returns generated by these indices. This is done using a GARCH-MIDAS framework that allows for the incorporation of data at different frequencies into the same model. The findings show that certain market and macroeconomic variables significantly affect the volatility of total returns generated by the constructed indices. Specifically, the JSE All-Share index is shown to account for approximately one third of corporate bond market volatility in the long run, which would suggest that monitoring performance at the aggregate equity index level could potentially improve the robustness of client portfolios holding corporate bond assets. The final substantive chapter builds on these conclusions by investigating the relative informational efficiency between the corporate bond and equity markets. This is done by using a VAR model to analyse the lead-lag relationship between bonds and equity issued by the same firms. The results are largely consistent with previous contributions to the literature showing that equity markets are relatively more informationally efficient and tend to lead corporate bond markets. This would suggest that equity returns are potentially predictive of future corporate bond returns and may allow professional investors to more appropriately manage risk in multi-asset portfolios.
  • Thumbnail Image
    Item
    Open Access
    A post-crisis investigation in to the performance of GARCH-based historical & analytical value-at-risk on the FTSE
    (2013) De Alessi, Alessando; Huang, Chun-Sung
    This paper is an investigation into the performance of GARCH-based VaR models on the South African FTSE/JSE Top 40 Index. Specifically, this paper investigates whether stability has returned to the VaR measure following its poor performance during the latest global financial crisis (2007). GARCH models are used in both an analytic and historical approach for modeling 1%, 2.5% and 5% daily VaR for a three year backtest period (2010-2012). Four distributions are used: the normal, generalised error, t-distribution and the skewed t-distribution. A particular question asked by this paper, is whether the data from the latest financial crisis (2007) should be used in estimating VaR in a post-crisis market. To investigate this, all models are re-estimated using data that has the financial crisis and/or high volatility period removed, then the results across the two data sets are compared. The take away point from this research is that the volatility-clustering mechanism inherent in every GARCH model is capable of producing accurate VaR estimates in a post-downturn/lower-volatility market even when the data on which the model was estimated contains financial downturn/volatile data. There is strong evidence suggesting stability has returned to this measure - however caution remains over using over-simplified models.
  • Thumbnail Image
    Item
    Open Access
    Risk parity and other risk based portfolio allocation approaches in South African and international equity markets
    (2014) Panulo, Barry; Van Rensburg, Paul; Huang, Chun-Sung
    Risk parity, a portfolio allocation technique based on the equalization of constituent risk contributions, has garnered significant attention in academic circles over the past decade. This study employs back-tests to explore the empirical performance of the approach relative to other prominent heuristic and risk based allocation techniques on South Africa's All Share Index (ALSI) and 12 auxiliary international equity indices. We find that the technique discharges its core risk contribution equalization objectives well in out of sample testing but appears to lag other risk based allocation techniques in terms of risk and return performance. We also establish links between the approaches' performance and leverage aversion theory and find some evidence that levels of market concentration may impact the performance of risk parity portfolios across equity indices.
  • No Thumbnail Available
    Item
    Open Access
    Using deep learning to characterise weak signals in global equity markets: a case study of COVID-19
    (2024) Clarke, Keegan G; Huang, Chun-Sung
    This study examines the use of deep learning to identify and characterise anomalous events and their preceding weak signals in equity price data. Particular interest is placed on Gray Rhino events, indicated by the presence of progressively stronger signals prior. The market behaviour prior to and during the COVID-19 pandemic on G-20 equity markets provides a useful context to this end. Existing literature has examined the effects of the pandemic on these markets but has yet to provide conclusive insights into the development of the major equity crash. We compare the existing literature concomitantly to our rigorous application of event study methodology, identifying the presence and effects of signals prior to the market crash. In addition, we develop and deploy a novel Anomaly Characterisation Process (ACP). The ACP utilises an ARIMA time series model to transform equity price time-series for the extraction of relevant information, whereby subsequent fits of GJR-GARCH and deep-undercomplete-autoencoder models are deployed. Resultantly, measures of dispersion and atypicality are produced which allow for effective and clear characterisation of the degree of typicality of the equity prices and their movements. This innovative method demonstrates efficacy in detecting both point and contextual anomalies. When applied in the context of COVID-19, the findings suggest that different event types can be distinguished successfully with this novel approach through the identification of weak signals. Notably, these insights of the ACP in conjunction with those of the event study suggest that the COVID-19 market crash is consistent with a Gray Rhino event and not a Black Swan event. We briefly demonstrate that these insights can be used by market participants to improve risk-adjusted returns via ACP-informed risk-mitigation techniques.