Thesis (MComm (Statistics and Actuarial Science))--University of Stellenbosch, 2006. / An overview of modern and historical interest rate model theory is given with the specific aim of derivative pricing. A variety of stochastic interest rate models are discussed within a South African market context. The various models are compared with respect to characteristics such as mean reversion, positivity of interest rates, the volatility structures they can represent, the yield curve shapes they can represent and weather analytical bond and derivative prices can be found. The distribution of the interest rates implied by some of these models is also found under various measures. The calibration of these models also receives attention with respect to instruments available in the South African market. Problems associated with the calibration of the modern models are also discussed.

This thesis is a technical inquiry into remedies for high inflation. In its center there is the usual tradeoff between inflation aversion on the one hand and some benefit from inflation via Phillips curve effects on the other hand. Most remarkable and pioneering work for us is the famous Barro-Gordon model - see (Barro & Gordon 1983a) respectively (Barro & Gordon 1983b). Parts of this model form the basis of our work here. Though being well known the discretionary equilibrium is suboptimal the question arises how to overcome this. We will introduce four different models, each of them giving a different perspective and way of thinking. Each model shows a (sometimes slightly) different way a central banker might deliver lower inflation than the one shot Barro-Gordon game at a first glance would suggest. To cut a long story short we provide a number of reasons for believing that the purely discretionary equilibrium may be rarely observed in real life. Further the thesis provides new insights for derivative pricing theories. In particular, the potential role of financial markets and instruments will be a major focus. We investigate how such instruments can be used for monetary policy. On the contrary these financial securities have strong influence on the behavior of the central bank. Taking this into account in chapters 3 and 4 we come up with a new method of pricing inflation linked derivatives. The latter to the best of our knowledge has never been done before - (Persson, Persson & Svenson 2006), as one of very view economic works taking into account financial markets, is purely focused on the social planer's problem. A purely game theoretic approach is done in chapter 2 to change the original Barro-Gordon. Here we deviate from a purely rational and purely one period wise thinking. Finally in chapter 5 we model an asymmetric information situation where the central banker faces a trade off between his current objective on the one hand and benefit arising from not perfectly informed agents on the other hand. In that sense the central bank is also concerned about its reputation.

This study estimates the interdependence between South Africa‟s monetary policy and stock market performance, utilising structural vector autoregression (SVAR) methodology. The study finds that a stock price shock which decrease stock prices by 100 basis points leads to 5 basis points decrease in interbank rate. A monetary policy shock that increases the interbank rate by l percent leads to decrease in real stock prices by 1 percent. This result for South Africa is similar to the result by Bjornland and Leteimo (2009) which earlier concluded that there was a high interdependence between interest rate setting and stock prices. However the magnitude of the relationship is relatively lower for South Africa compared to that of the United States of America (USA). The result of the current study is also very much consistent with the argument that the South African stock market is resource-based and so is influenced by external shocks, meaning monetary policy shock does not have as much impact on stock market in South Africa as in the USA. However the SARB may have to consider watching movements in stock prices so that booms in stock markets do not defeat central bank monetary policy thrusts. The stock price market is an essential source of information for monetary policy in South Africa.

There is a growing number of applications of machine learning and deep learning in quantitative and computational finance. In this thesis, we focus on two of them. In the first application, we employ machine learning and deep learning in derivative pricing. The models considering jumps or stochastic volatility are more complicated than the Black-Merton-Scholes model and the derivatives under these models are harder to be priced. The traditional pricing methods are computationally intensive, so machine learning and deep learning are employed for fast pricing. I n Chapter 2, we propose a method for pricing American options under the variance gamma model. We develop a new fast and accurate approximation method inspired by the quadratic approximation to get rid of the time steps required in finite difference and simulation methods, while reducing the error by making use of a machine learning technique on pre-calculated quantities. We compare the performance of our method with those of the existing methods and show that this method is efficient and accurate for practical use. In Chapters 3 and 4, we propose unsupervised deep learning methods for option pricing under Lévy process and stochastic volatility respectively, with a special focus on barrier options in Chapter 4. The unsupervised deep learning approach employs a neural network as the candidate option surface and trains the neural network to satisfy certain equations. By matching the equation and the boundary conditions, the neural network would yield an accurate solution. Special structures called singular terms are added to the neural networks to deal with the non-smooth and discontinuous payoff at the strike and barrier levels so that the neural networks can replicate the asymptotic behaviors of options at short maturities. Unlike supervised learning, this approach does not require any labels. Once trained, the neural network solution yields fast and accurate option values. The second application focuses on financial time series simulation utilizing deep learning techniques. Simulation extends the limited real data for training and evaluation of trading strategies. It is challenging because of the complex statistical properties of the real financial data. In Chapter 5, we introduce two generative adversarial networks, which utilize the convolutional networks with attention and the transformers, for financial time series simulation. The networks learn the statistical properties in a data-driven manner and the attention mechanism helps to replicate the long-range dependencies. The proposed models are tested on the S&P 500 index and its option data, examined by scores based on the stylized facts and are compared with the pure convolutional network, i.e. QuantGAN. The attention-based networks not only reproduce the stylized facts, including heavy tails, autocorrelation and cross-correlation, but also smooth the autocorrelation of returns.

This research tests for persistence in tax generated earnings changes. Earnings persistence is indicated by the capitalization of earnings by securities markets. This research disaggregates accounting earnings and examines the security markets’ evaluation of the relative permanence or transience of the component of earnings resulting from revenue law changes. Two proxies for tax generated earnings changes are evaluated though an examination of earnings response coefficients. The results indicate that tax generated earnings changes are not expected to persist beyond two accounting periods, and may reflect the ability of firms to manage tax earnings. / Ph. D.

Submitted by Elba Lopes (elba.lopes@fecap.br) on 2017-11-18T16:34:14Z No. of bitstreams: 2 ANATOLI DIAS ORLOVAS.pdf: 380790 bytes, checksum: ed39ab77e72e71915d8186a36ca1276b (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-11-18T16:34:14Z (GMT). No. of bitstreams: 2 ANATOLI DIAS ORLOVAS.pdf: 380790 bytes, checksum: ed39ab77e72e71915d8186a36ca1276b (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-03-28 / The objective of this study is to verify if the use of multiples for valuation of companies during their initial public offering is a good predictor of their stock prices. For this study, 71 IPO???s from the Brazilian market that met the selection criteria were analyzed among 117 IPO???s that were approved by the CVM during the period from 2004 to 2014, using the three valuation multiples with higher incidence of recommendations among the analyzed scientific studies surveyed for the elaboration of this work: Price to Book, Price to Earnings and Enterprise Value to EBITDA. This study has compared the multiples of IPOs with the same multiples of companies belonging to the same sector of activity (peers), in two distinct moments: (i) the day of the IPO's and (ii) the day before of the IPO. The analysis was structured in three models: (i) relating the multiple of the IPO to the mean, median and harmonic mean of this same multiple for their peers, (ii) adding dummies related to the year of IPO in the previous model and (iii) adding variables related to size, leverage and return to the previous model. For all models, MQO regressions were performed. As a result, the work found statistical significance to the multiple Price to Earnings when the mean was used in the Model III, showing significance with the use of the difference of returns. Statistical significance was also founded to the multiple Enterprise Value to EBITDA when the mean was used to the Model I, which show the best adjustments demonstrated by R??, although it did not find statistical significance for the Price to Book multiple. / O trabalho a seguir tem por objetivo analisar se a utiliza????o de m??ltiplos para a avalia????o de empresas no momento de sua abertura de capital ?? um bom previsor para o pre??o das a????es destas empresas. Para este trabalho foram analisados 71 IPO???s do mercado brasileiro que atenderam aos crit??rios de sele????o entre os 117 IPO???s que foram aprovadas pela Comiss??o de Valores Mobili??rios ??? CVM, durante o per??odo de 2004 a 2014, utilizando-se dos tr??s m??ltiplos com maior incid??ncia de recomenda????es entre os estudos pesquisados para a elabora????o deste trabalho: Price to Book, Price to Earnings e Enterprise Value to EBITDA. O trabalho comparou os m??ltiplos dos IPO???s com os mesmos m??ltiplos de empresas pertencentes ao mesmo setor de atua????o (peers), em dois momentos distintos: (i) o dia da estreia do IPO na bolsa e (ii) o dia de negocia????es anterior a esta estreia. As an??lises foram estruturadas em 3 modelos, sendo: (i) relacionando o m??ltiplo do IPO com a m??dia, mediana e m??dia harm??nica deste mesmo m??ltiplo para seus peers, (ii) acrescentando dummies referente ao ano da emiss??o do IPO ao modelo anterior e (iii) acrescentando vari??veis relacionadas ao tamanho, alavancagem e endividamento ao modelo anterior. Para todos os modelos, foram realizadas regress??es MQO. Como resultado, o trabalho encontrou signific??ncia estat??stica para os m??ltiplos Price to Earnings quando utilizada a m??dia para o Modelo III, apresentando signific??ncia com a utiliza????o da diferen??a dos retornos. Encontrou tamb??m signific??ncia estat??stica para o m??ltiplo Enterprise Value to EBITDA, quando utilizada a m??dia para o Modelo I, o qual apresentou melhores ajustes demonstrados pelo R??, embora n??o tenha encontrado signific??ncia estat??stica para o m??ltiplo Price to Book.

Thesis (PhD)--Macquarie University, Division of Economic and Financial Studies, Dept. of Statistics, 2008. / Bibliography: p. 73-77. / Introduction -- Lévy processes used in option pricing -- Option pricing for Lévy processes -- Option pricing based on empirical characteristic functions -- Performance of the five models on historical data -- Conclusions -- References -- Appendix A. Proofs -- Appendix B. Supplements -- Appendix C. Matlab programs. / Pricing problems of financial derivatives are among the most important ones in Quantitative Finance. Since 1973 when a Nobel prize winning model was introduced by Black, Merton and Scholes the Brownian Motion (BM) process gained huge attention of professionals professionals. It is now known, however, that stock market log-returns do not follow the very popular BM process. Derivative pricing models which are based on more general Lévy processes tend to perform better. --Carr & Madan (1999) and Lewis (2001) (CML) developed a method for vanilla options valuation based on a characteristic function of asset log-returns assuming that they follow a Lévy process. Assuming that at least part of the problem is in adequate modeling of the distribution of log-returns of the underlying price process, we use instead a nonparametric approach in the CML formula and replaced the unknown characteristic function with its empirical version, the Empirical Characteristic Functions (ECF). We consider four modifications of this model based on the ECF. The first modification requires only historical log-returns of the underlying price process. The other three modifications of the model need, in addition, a calibration based on historical option prices. We compare their performance based on the historical data of the DAX index and on ODAX options written on the index between the 1st of June 2006 and the 17th of May 2007. The resulting pricing errors show that one of our models performs, at least in the cases considered in the project, better than the Carr & Madan (1999) model based on calibration of a parametric Lévy model, called a VG model. --Our study seems to confirm a necessity of using implied parameters, apart from an adequate modeling of the probability distribution of the asset log-returns. It indicates that to precisely reproduce behaviour of the real option prices yet other factors like stochastic volatility need to be included in the option pricing model. Fortunately the discrepancies between our model and real option prices are reduced by introducing the implied parameters which seem to be easily modeled and forecasted using a mixture of regression and time series models. Such approach is computationaly less expensive than the explicit modeling of the stochastic volatility like in the Heston (1993) model and its modifications. / Mode of access: World Wide Web. / x, 111 p. ill., charts

It is generally accepted in academic circles that the developed country capital markets with their advanced infra-structure, depth and liquidity are at a minimum Weak-Form efficient. Since the Weak-Form EMH proposes that current security prices immediately assimilate all historical information, it therefore also implies that technical analysis (which relies on charts and analysis of past price patterns to extrapolate future price movements) would be a futile exercise. Yet technical analysis has endured over time and is still an intensively and widely used investment analysis technique. This indicates a clear disconnect between technical analysis as employed by practitioners in the market and the technical analysis methodologies utilized by academics in prior Weak-Form EMH studies. The problem is prior technical analysis Weak-Form EMH studies were burdened with methodological weaknesses which severely handicapped the profit generating potential of technical analysis and suggest that previous Weak-Form EMH research findings were erroneous in being unable to reject the null Weak- Form market efficiency hypothesis. This study addresses the problem by eliminating prior methodological weaknesses and utilizing high frequency intra-day data, the combination of qualitative and quantitative techniques and volume signals to develop a portfolio of Intermarket Momentum technical analysis strategies that generate significant excess profits. The objective of this study is therefore to provide evidence that contrary to prior research findings, the developed country capital markets are not Weak-Form efficient. The results show that the portfolio of Intermarket Momentum trading strategies generated returns in excess of the market with a significantly positive Alpha of 8.52% that allowed the rejection of the Null Hypothesis and the acceptance of the Alternative Hypothesis that the developed country capital markets are not Weak-Form efficient, thereby refuting the widely accepted EMH. / Business Management / D.B.L.