Orientador: Luiz Koodi Hotta / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-23T04:22:11Z (GMT). No. of bitstreams: 1 Almeida_Danielde_M.pdf: 17481253 bytes, checksum: 669620c3fe4155707f86370dd1778d01 (MD5) Previous issue date: 2013 / Resumo: O objetivo da dissertação é estudar modelos de volatilidade que consideram dois tipos de assimetria usualmente encontradas em séries de finanças, a assimetria das perturbações e o efeito de alavancagem. Perturbações assimétricas são utilizadas devido ao fato estilizado de que perdas têm distribuição com cauda mais pesada do que ganhos. Já o efeito de alavancagem leva em consideração que perdas têm maior influência na volatilidade do que os ganhos. São estudados os modelos GARCH univariados que contemplam os dois tipos de assimetria separadamente e conjuntamente e modelos GARCH multivariados que permitem o efeito de alavancagem. Os resultados são apresentados em dois artigos. O primeiro descreve os principais modelos univariados que possam explicar estes dois fatos estilizados e analisa, com detalhes, oito séries: os índices Ibovespa, Merval e S&P 500, e as ações Itaú-Unibanco, Vale, Petrobras, Banco do Brasil e do Bradesco. A conclusão é que os dois tipos de assimetria estão presentes nas séries, na maioria das vezes simultaneamente. O segundo artigo faz uma revisão dos principais modelos multivariados da família GARCH, incluindo modelos com efeitos assimétricos nas variâncias e nas covariâncias condicionais. Alguns destes modelos são analisados com mais detalhes através de simulações. Considerou-se as perdas de eficiência na estimativa da matriz de volatilidade ao se ter erros de especificação, isto é ajustar um determinado modelo a séries geradas por outros modelos. Os modelo mais utilizados na literatura são aplicados a uma série trivariada, contendo o índice Ibovespa e as ações Petrobras e Vale. Os três modelos selecionados pelos critérios AIC e BIC, possuem o efeito de alavancagem / Abstract: The objective of this dissertation is to study volatility models that consider two types of asymmetry usually found in finance series, the skewness of the innovations and the leverage effect. Skewness means that the distribution of losses has a heavier tail than the distribution of gains. The leverage effect stems from the fact that losses have a greater influence on future volatilities than gains. It is considered univariate GARCH models that include both types of asymmetry, separately and jointly, and multivariate GARCH models that allow for leverage effects. The results are presented in two papers. The first one describes the main univariate models that consider these two stylized facts and analyzes, in detail, eight series: the Ibovespa, Nasdaq and S&P 500 indices, and the Itaú-Unibanco, Vale, Petrobras, Banco do Brasil and Bradesco stocks. The conclusion is that both stylized facts are present in some series, mostly simultaneously. The second paper reviews the main multivariate GARCH models, including models with asymmetric effects on conditional variances and covariance. Some of these models are analyzed in more detail through simulations. The most used models in the literature are applied to a three-dimensional time series, containing the Bovespa index and the Petrobras and Vale markets. The three models selected by AIC and BIC criteria allow for leverage effects / Mestrado / Estatistica / Mestre em Estatística

My dissertation investigated business cycle effects on US sectoral stock returns. The first chapter examined the relationship between the business cycle and sectoral stock returns. First, I calculated constant correlation coefficients between the business cycle and sectoral stock returns. Then, I employed the DCC GARCH model to estimate time-varying correlation coefficients for each pair of the business cycle and sectoral stock returns. Finally, I ran regression of sectoral returns on dummy variables designed to capture the four stages of the business cycle. I found that though sectoral stock returns were closely related to the business cycle, they did not share some of its main characteristics. The second chapter developed two models in order to discuss possible asymmetric business cycle effects on US sectoral stock returns. One was a GARCH model with asymmetric explanatory variables and the other one was an ARCH-M model with asymmetric external regressors. In the second model, square root of conditional variance of the business cycle proxy was characterized as positive or negative risk, depending on the algebraic sign of past innovations driving the business cycle proxy. I found that some sectors changed their cyclicities from expansions to recessions. Negative shocks to business cycles had most power to influence sectoral volatilities. Positive and negative parts of business cycle risk had same effects on some sectors but had opposite effects on other sectors. A general conclusion of both models was that business cycle had stronger effects than own sectoral effects in driving sectoral returns. The third chapter discussed Chinese business cycle effects on US sectoral stock returns at two horizons. At a monthly horizon, the third lag of Chinese IP growth rate had positive effects on most sectors. The second lag of US IP growth rate had positive effects on almost all sectors. At a quarterly horizon, besides the extensive positive effects of the first lag of Chinese IP growth rate, the third and fourth lags also had effects on some sectors. The US IP growth rate had the same pattern, namely positive first and fourth lag effects and negative third lag effects. Using a 5-year rolling fixed window, I found that these business cycle effects were time-varying. The major changes in parameters resulted from the elimination of quota on textiles by WTO, the terrorist attacks on the US, and the 2007 financial crisis.

The main goal of this thesis is to suggest an appropriate approach to interest rate risk modeling in the Czech financial market environment in various situations. Three distinct periods are analyzed. These periods, which are the period before the global financial crisis, period during the financial crisis and in the aftermath of the global financial crisis and calming subsequent debt crisis in the eurozone, are characterized by different evaluation of liquidity and credit risk, different relationship between financial variables and market participants and different degree of market regulations. Within this goal, an application of the BGM model in the Czech financial market environment is crucial. Use of the BGM model for the purpose of predicting a dynamics of a yield curve is not very common. This is firstly due to the fact that primary use of this model is a valuation of interest rate derivatives while ensuring the absence of arbitrage and secondly its application is relatively difficult. Nevertheless, I apply the BGM model to obtain predictions of the probability distributions of interest rates in the Czech and eurozone market environment, because its complexity, direct modeling of a yield curve based on market rates and especially a possibility of parameter estimation based on current swaptions volatilities quotations may lead to a significant improvement of predictions. This improvement was also confirmed in this thesis. Use of swaptions volatilities market quotations is especially useful in the period of unprecedented mone- tary easing and increased number of central banks and other regulators interventions into financial markets that occur after the financial crisis, because it reflects current market expectations which also include future interventions. As a consequence of underdevelopment of the Czech financial market there are no market quotations of Czech koruna denominated swaptions volatilities. I suggest their approximations based on quotations of euro denominated swaptions volatilities and also using volatilities of koruna and euro forward rates. Use of this approach ensures that predictions of the Czech yield curve dynamics contain current market expectations. To my knowledge, any other author has not presented similar application of the BGM model in the Czech financial market environment. In this thesis I further predict a Czech and Euro area money market yield curve dynamics using the CIR and the GP models as representatives of various types of interest rates models to compare these predictions with BGM predictions. I suggest a comprehensive system of three criteria, based on comparison of predicti- ons with reality, to describe a predictive power of selected models and an appropria- teness of their use in the Czech market environment during different situations in the market. This analysis shows that predictions of the Czech money market yield curve dynamics based on the BGM model demonstrate high predictive power and the best 8 quality in comparison with other models. GP model also produces relatively good qua- lity predictions. Conversely, predictions based on the CIR model as a representative of short rate model family completely failed when describing reality. In a situation when the economy allows negative rates and there is simultaneously a significant likelihood of their implementation, I recommend to obtain predictions of Czech money market yield curve dynamics using GP model which allows existence of negative interest rates. This analysis also contains a statistical test for validating the predictive power of each model and information on other tests. Berkowitz test rejects a hypothesis of accurate predictions for each model. However, this fact is common in real data testing even when using relatively good model. This fact is especially caused by difficult fulfilment of test conditions in real world. To my knowledge, such an analysis of the predictive power of selected interest rate models moreover in the Czech financial market environment has not been published yet. The last goal of this thesis is to suggest an appropriate approach to obtaining pre- dictions of Czech government bonds risk premium dynamics. I define this risk premium as a difference between government bond yields and fixed rate of CZK IRS with the same length. I apply the GP model to describe the dynamics of this indicator of the Czech Republic credit risk. In order to obtain a time series of the risk premium which are necessary for estimation of GP model parameters I firstly estimate yield curves of Czech government bonds using Svensson model for each trading day since 2005. Resulting si- mulations of risk premium show that the GP model predicts the real development of risk premiums of all maturities relatively well. Hence, the proposed approach is suitable for modeling of Czech Republic credit risk based on the use of information extracted from financial markets. I have not registered proposed approach to risk premium modeling moreover in the Czech financial market environment in other publications.

Thesis ( M.Sc.(Agricultural Economics)) --University of Limpopo, 2019. / Potato is perceived as an excellent crop in the fight against hunger and poverty. The recent high potato price in South Africa has pushed the vegetable out of reach of the poorest of the poor. The study attempts to analyse potato price volatility in South Africa and furthermore assess how various factors were responsible for the recent potato price volatility. Quarterly data for potato price, number of hectares planted, rainfall and temperature levels from 2006q1 to 2017q4 was collected from various sources and were used for analysis. The total observation of 48. The volatility in the series was determined by performing ARCH/GARCH model. GARCH model indicates an evidence of GARCH effect in the series, meaning that GARCH model influences potato price volatility in South Africa. The Johansen cointegration used both trace and eigenvalue to test the existence of a long run relationship between potato price and various variables. The cointegration results were positive indicating that there exists long run relationship amongst variables. The study further used Johansen cointegration as well as standard error to determine the number of cointegrating variables in the long run. The results indicated that the number of hectares planted and rainfall level have significant relationship with potato price. Wald tests was used to check whether the past values of number of hectares planted and rainfall level influenced the current value of potato price. The Walt test results concluded that there is no evidence of short run causality running from number of hectares planted and rainfall level to potato price. In the study, ECM model was used to forecast the potato price fluctuation in South Africa. The study recommends that farmers need to engage in contract market so as to minimize the risk of potato price volatility. The Department of Agriculture should forecast agricultural commodities price volatility and make information accessible to the farmers so that they are able to adopt strategies that will assist them to overcome crisis.

In this thesis the GARCH models are applied to evaluate financial options and futures. In the first application, the GARCH models in parsimonious form are studied for pricing the S&P500 options. Unlike previous studies that focus on developed formulation, the results indicate that simplified models provide effective performance and it is the simple GARCH model that yields the least valuation error. To our consideration, examining model possessing simplification is of practical importance because model estimation becomes readily accessible through available econometric software, which circumvent programming barriers in implementing alternative one’s own pricing methods. The second application consider the component GARCH models for currency option pricing. The valuation results favour the component formulations particularly in the pricing of long term contracts. Volatility modelling results indicate that the return-volatility relationship is symmetric in the long run, but over the short term asymmetry also arises in the EURUSD and GBPUSD exchange rates. The third application evaluates canola futures in Canada in relation to spot market price. Results confirm the cointegrating relationship with threshold corresponding to transaction and adjustment cost. And it is the futures market that adjusts actively to price disparities but in the meantime there is volatility spillover from futures to the spot market. Overall, our empirical assessments indicate the importance of the time varying volatility and the improvements achieved in option pricing and futures evaluation. We believe the present study’s analysis provides useful suggestions and further guidance to practitioners and investors for the pricing and trading in the equity and foreign exchange markets, also to the market agents to better evaluate price uncertainty in order to guard against adverse price changes.

We propose new estimation methods for the factor loading matrix in modeling multivariate volatility processes. The key step of the methods is based on the weighted scatter estimators, which does not involve optimizing any objective function and was embedded with robust estimation properties. The method can therefore be easily applied to high-dimensional systems without running into computational problems. The estimation is proved to be consistent and the asymptotic distribution is derived. We compare the performance with other estimation methods and demonstrate its superiority when using both simulated data as well as real-world case studies. / Statistics

Doctor of Philosophy / Department of Economics / Lance J. Bachmeier / This dissertation investigates the long run and short run relationships between Saudi stock market returns and eight macroeconomic variables. We investigate the ability of these variables to predict the level and volatility of Saudi stock market returns. A wide range of Vector autoregression (VAR) and generalized autoregressive conditional heteroskedasticity (GARCH) models estimated and interpreted. A Johansen-Juselius cointegration test indicates a positive long run relationship between the Saudi stock price index and the M2 money supply, bank credit, and the price of oil, and a negative long run relationship with the M1 money supply, the short term interest rate, inflation, and the U.S. stock market. An estimated vector error correction model (VECM) suggests significant unidirectional short run causal relationships between Saudi stock market returns and the money supply and inflation. The VECM also finds a significant long run causal relationship among the macroeconomic variables in the system. The estimated speed of adjustment indicates that the Saudi stock market converges to the equilibrium within half a year. Granger causality tests show no causal relationship between Saudi stock market returns and the exchange rate. Impulse response function analysis shows no significant relationship between Saudi stock market returns and the macroeconomic variables. Forecast error variance decompositions suggest that 89% of the variation in Saudi stock market returns is attributable to its own shock, which implies that Saudi stock market returns are largely independent of the macroeconomic variables in the system. Finally, a GARCH-X model indicates a significant relationship between volatility of Saudi stock returns and short run movements of macroeconomic variables. Implications of this study include the following. (i) Prediction of stock market returns becomes more difficult as the volatility of the macroeconomic variables increases in the short run. (ii) Investors should look at the systematic risks revealed by these macroeconomic variables when structuring their portfolios and diversification strategies. (iii) Policymakers should seek to minimize macroeconomic fluctuations considering the effect of macroeconomic variables changes on the stock market when formulating economic policy.

American Depository Receipts (ADRs) represent shares of foreign firms that are issued and traded in the U.S. Since an ADR and its underlying shares represent ownership interest of the same firm, they should be perfect substitutes in a perfect market. However, market imperfections such as differences in information environment, liquidity, investment and trading restrictions, taxes, control right, corporate governance might make them less-than-perfect substitutes. These imperfections, on the other hand, also present opportunities for research. This dissertation consists of two essays on ADRs, both related to the effects of less-thanperfect information. Specifically, the first essay examines the return and volatility transfers between ADRs and their underlying home shares. Our investigation differs from the previous studies in that we cover substantially more countries and that we attempt to explain the variations in the extents of transfer effects both across firms and across countries. Various hypotheses are developed, based on the premise that barriers associated with trading, investments, and corporate governance would lower the extent or effectiveness of transfers. Overall, our empirical results support these hypotheses. The second essay takes the viewpoint of the issuing firms. Supposedly, an issuer's timing and dollar amount raised depend on the conditions of three markets: its home equity market, the U.S. equity market, and the currency market. From purely the standpoint of information accessibility, ADR issuers are likely to time their issues or set their amounts with respect to the conditions of the home equity market and/or currency market, with which they are more familiar. On the other hand, issuers typically employ the assistance of U.S. investment banks, and therefore they may be well-informed about the U.S. equity market. This is largely an empirical issue. Generally, our empirical results are mixed, but there is somewhat stronger evidence for the U.S. equity market being more important when setting the issue amount. There is also evidence that suggests regulations having influences on such activities.

Background: Stress tests have been increasingly used as a part of the supervisory tool by national regulators after the financial crisis, which can also be used to conduct authorities’ supervisory for determining bank capital levels, assessing the health of a bank. Purpose: The main purpose of this study is to assess whether some micro factors play important roles on the changes of Common Equity Tier One Capital Ratio (between the bank accounting value and the stress testing results under the adverse scenarios).  Our secondary purpose is to investigate if our empirical results will help to provide some theoretical suggestions to regulators when they exercise stress tests.   Method: An empirical analysis by using Panel Data, introducing GARCH model to measure volatility.   Empirical foundation: The results of EU-wide stress tests and bank financial statements   Conclusion: The coefficient associated with non-performing loans to total loans is positively significant and the coefficient associated with bank size is negatively significant.  In addition, the financial system of strong banks is better to absorb financial shocks. These findings are useful, as banks is a reflection of the financial stability of an economic entity, we can use these findings as another reason to pay attention to the process of the stress testing rather just stress testing results.

[pt] O modelo implementado neste trabalho, proposto em Barone-Adesi, Engle e Mancini (2008), utiliza o método da Simulação Histórica Filtrada em conjunto com a simulação de Monte Carlo para calibração de parâmetros de um modelo GARCH a partir do qual opções do mercado brasileiro são apreçadas. Os retornos da simulação são gerados a partir das inovações empíricas obtidas no modelo GARCH assimétrico ajustado aos retornos diários das ações. Os resultados obtidos apontam para ajustes satisfatórios dentro da amostra, quando comparado ao modelo de Black E Scholes. No entanto, fora da amostra, resultados similares foram verificados para ambos os modelos de apreçamento. / [en] The model implemented in this work, proposed by Barone-Adesi, Engle, and Mancini (2008), applies the Filtered Historical Simulation method based on Monte Carlo simulation to calibrate the parameters of a GARCH model in which options from Brazilian market are priced. The simulated returns are generated from empirical innovations obtained by an asymmetric GARCH model adjusted for daily stock returns. The results suggest a satisfactory in-sample fit when compared to the Black E Scholes model. However, similar results were observed out-of-sample for both pricing models.