No description available.

O presente trabalho analisou as volatilidades dos mercados de renda fixa e variável de onze países, sendo eles: Brasil, Rússia, Índia, China, África do Sul (neste país apenas renda fixa), Argentina, Chile, México, Estados Unidos, Alemanha e Japão no período de janeiro de 2000 a dezembro de 2011. Os indicadores utilizados para representar cada mercado foram os índices dos mercados de ações e as taxas de juros interbancárias. Para tanto, o estudo se utilizou de modelos de heterocedasticidade condicional auto-regressiva: ARCH, GARCH, EGARCH, TGARCH e PGARCH, verificando quais destes processos eram mais eficientes para modelagem da volatilidade dos mercados dos países da amostra. Esta pesquisa também verificou qual dos modelos (ARIMA ou modelos GARCH e suas extensões) conseguiria prever melhor as séries de tempo analisadas. Além disso, por meio dos índices de correlação, covariância e causalidade Granger, foram comparados os retornos e a volatilidade do mercado de ações entre os países BRIC, entre os países latinos americanos e entre os países desenvolvidos e o Brasil. Os resultados sugerem que a volatilidade, tanto do mercado de renda fixa quanto do mercado de renda variável, é mais bem modelada por processos GARCH assimétricos (EGARCH e TGARCH), demonstrando efeitos de alavancagem nas séries estudadas. Quanto aos modelos de previsão, os modelos ARIMA, também para os dois mercados, mostrou-se mais eficiente que os modelos GARCH e suas extensões. Além disso, as volatilidades dos mercados de ações entre os países analisados parecem ser mais correlacionadas e possuir maior causalidade Granger do que os retornos destes países. Entre os dois mercados, renda fixa e variável dentro de cada país, as correlações dos retornos e da volatilidade são muito baixas, em algumas vezes negativa, e há pouca relação de causalidade Granger. / This study analyzed the volatility of fixed income and stocks markets for eleven countries, namely: Brazil, Russia, India, China, South Africa (just fixed income), Argentina, Chile, Mexico, United States, Germany and Japan from January 2000 to December 2011, using interbank interest rate as a fixed income market indicator and stock index to each country, as a stock market indicator. Therefore, the study used models of autoregressive conditional heteroscedasticity: ARCH, GARCH, EGARCH, TGARCH e PGARCH to verify which of these processes were more effective for in volatility modeling in each country. This research also found that the models (ARIMA or GARCH models and their extensions) could be used as the best forecast models. Moreover, by means of correlation coefficients, covariance and Granger causality, were used to compare the returns and volatility of the stock market among the BRIC countries, among the Latin American countries and between developed countries and Brazil. The results suggest that the volatility of both the fixed income market as the stock market is best modeled by processes asymmetric GARCH (EGARCH and TGARCH) demonstrating leverage effects in the time series. Regarding prediction ARIMA models was more efficient for both markets than GARCH models and extensions. In addition, the volatility of stock markets across countries analyzed seem to be more correlated and have higher Granger causality than returns these countries. Between the two markets, for each country, the correlations of returns and volatility are very low, if not positive, and there is low Granger causality.

O presente trabalho analisou as volatilidades dos mercados de renda fixa e variável de onze países, sendo eles: Brasil, Rússia, Índia, China, África do Sul (neste país apenas renda fixa), Argentina, Chile, México, Estados Unidos, Alemanha e Japão no período de janeiro de 2000 a dezembro de 2011. Os indicadores utilizados para representar cada mercado foram os índices dos mercados de ações e as taxas de juros interbancárias. Para tanto, o estudo se utilizou de modelos de heterocedasticidade condicional auto-regressiva: ARCH, GARCH, EGARCH, TGARCH e PGARCH, verificando quais destes processos eram mais eficientes para modelagem da volatilidade dos mercados dos países da amostra. Esta pesquisa também verificou qual dos modelos (ARIMA ou modelos GARCH e suas extensões) conseguiria prever melhor as séries de tempo analisadas. Além disso, por meio dos índices de correlação, covariância e causalidade Granger, foram comparados os retornos e a volatilidade do mercado de ações entre os países BRIC, entre os países latinos americanos e entre os países desenvolvidos e o Brasil. Os resultados sugerem que a volatilidade, tanto do mercado de renda fixa quanto do mercado de renda variável, é mais bem modelada por processos GARCH assimétricos (EGARCH e TGARCH), demonstrando efeitos de alavancagem nas séries estudadas. Quanto aos modelos de previsão, os modelos ARIMA, também para os dois mercados, mostrou-se mais eficiente que os modelos GARCH e suas extensões. Além disso, as volatilidades dos mercados de ações entre os países analisados parecem ser mais correlacionadas e possuir maior causalidade Granger do que os retornos destes países. Entre os dois mercados, renda fixa e variável dentro de cada país, as correlações dos retornos e da volatilidade são muito baixas, em algumas vezes negativa, e há pouca relação de causalidade Granger. / This study analyzed the volatility of fixed income and stocks markets for eleven countries, namely: Brazil, Russia, India, China, South Africa (just fixed income), Argentina, Chile, Mexico, United States, Germany and Japan from January 2000 to December 2011, using interbank interest rate as a fixed income market indicator and stock index to each country, as a stock market indicator. Therefore, the study used models of autoregressive conditional heteroscedasticity: ARCH, GARCH, EGARCH, TGARCH e PGARCH to verify which of these processes were more effective for in volatility modeling in each country. This research also found that the models (ARIMA or GARCH models and their extensions) could be used as the best forecast models. Moreover, by means of correlation coefficients, covariance and Granger causality, were used to compare the returns and volatility of the stock market among the BRIC countries, among the Latin American countries and between developed countries and Brazil. The results suggest that the volatility of both the fixed income market as the stock market is best modeled by processes asymmetric GARCH (EGARCH and TGARCH) demonstrating leverage effects in the time series. Regarding prediction ARIMA models was more efficient for both markets than GARCH models and extensions. In addition, the volatility of stock markets across countries analyzed seem to be more correlated and have higher Granger causality than returns these countries. Between the two markets, for each country, the correlations of returns and volatility are very low, if not positive, and there is low Granger causality.

In this work we want to examine selected models with nonlinear volatility and their properties. At the beginning we define models with non-constant variance, especially ARCH, GARCH and EGARCH models. Then we study the probability distributions that are mainly used in the EGARCH model. Then we focus on the EGARCH model, describe the conditions for stationarity and invertibility of the model, define diagnostic tests and QMLE estimates of parameters. In the last chapter we perform simulation studies of the selected models and their application to real data. 1

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-07-18T14:26:58Z No. of bitstreams: 1 daniellealdepaulaestevesdosreis.pdf: 1239258 bytes, checksum: 75cc07cdf6eba15d62c43b78ac783fbc (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-07-22T15:03:54Z (GMT) No. of bitstreams: 1 daniellealdepaulaestevesdosreis.pdf: 1239258 bytes, checksum: 75cc07cdf6eba15d62c43b78ac783fbc (MD5) / Made available in DSpace on 2016-07-22T15:03:54Z (GMT). No. of bitstreams: 1 daniellealdepaulaestevesdosreis.pdf: 1239258 bytes, checksum: 75cc07cdf6eba15d62c43b78ac783fbc (MD5) Previous issue date: 2011-12-16 / FAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais / Medidas de volatilidade se constituem numa preocupação por parte de estudiosos e profissionais do mercado financeiro. Modelos da família ARCH/GARCH a partir dos retornos diários produzem um indicador de volatilidade, mas, não conferem ao pesquisador uma medida observável do grau de variabilidade dos retornos em torno de seu valor esperado. A recente disponibilidade de dados de frequência inferior a um dia de negociação permitiu a elaboração de indicadores de volatilidade observáveis por meio de uma medida conhecida como volatilidade realizada. A partir de então, é possível elaborar um indicador observável de volatilidade diária com base em dados de natureza intradiária, de modo a representar uma medida mais apropriada do grau de risco de um ativo ou carteira de ativos, e, a partir de então, estimar a volatilidade por meio de processo da família ARIMA. De posse dos dados de alta-frequência de um papel preferencial da Petrobrás S.A., o presente trabalho se propõe, portanto, em construir a medida de volatilidade realizada por meio da soma dos quadrados dos retornos obtidos em intervalos regulares (5, 15 e 30 minutos) durante cada dia de negociação do papel PETR4 durante o período de 02/01/2007 à 29/10/2010. Posteriormente à criação do indicador de volatilidade realizada que se supõe como mais apropriado para se mensurar o grau de risco, pretende-se comparar a qualidade do ajustamento e a capacidade preditiva de cada um dos métodos de modelagem da volatilidade. A comparação dos modelos baseados em dados diários e intradiários dar-se-á por meio do cômputo do erro quadrático médio (EQM) e dos testes de Diebold e Mariano e de Harvey para avaliação da acurácia preditiva dos modelos. Os resultados mostraram que, em geral, os modelos da família ARIMA são mais apropriados para a avaliação do grau de ajustamento, e produz previsões mais satisfatórias que os modelos da família ARCH/GARCH. / Volatility measures constitute a concern among scholars and professionals of the financial market. Models of the ARCH/GARCH class from the daily returns produce an indicator of volatility, but do not give the researcher an observable measure of the degree of variability of returns around their expected value. The recent availability of data at frequencies below a trading day allowed the development of indicators of volatility observable through a measurement known as realized volatility. Since then, they can build an observable indicator of daily volatility based on intraday data, so as to represent a more appropriate measure of the riskiness of an asset, and from then estimate volatility through a process of ARIMA family. Provided with the data of a high frequency preferential role of Petrobrás S. A., the present paper therefore proposes to construct a measure of realized volatility by the sum of the squares of the returns obtained at regular intervals (5, 15 and 30 minutes ) during each trading day for the paper PETR4 during 02/01/2007 to 29/10/2010. After the creation of the realized volatility indicator that is supposed to be more appropriate to measure the degree of risk, the intent is to compare the goodness of fit and predictive ability of each of the methods of volatility’s models. The comparison of models based on daily data and intraday give will be through the calculation of the mean square error (MSE) and tests of Diebold and Mariano and Harvey to evaluate the predictive accuracy of models. The results in general showed that the models of the ARIMA class are more suitable for assessing the degree of adjustment and produces predictions more satisfactory than the models of the ARCH/GARCH class.

This thesis deals with various issues regarding determinants of US corporate credit spreads. These spreads are estimated as the difference between yields to maturity for corporate bonds and default-free instruments (Treasury bonds) of the same maturity. Corporate credit spreads are considered as measures of default risk. However, the premium required by investors for holding risky rather than risk-free bonds will incorporate a compensation not only for the default risk but also for other factors related to corporate bonds such as market liquidity or tax differential between corporate and Treasury bonds. In this study we firstly examine the relationship between bond ratings and credit spreads given that bond rating changes are expected to carry some informational value for debt investors. The findings indicate that bond ratings generally carry some informational value for corporate bond investors. The Granger causal relationship is more evident for negative watch lists and during periods of uncertainty in financial markets. In line with previous studies, our results suggest that changes in credit spreads are significantly related to interest rate levels, systematic risk factors (Fama and French) factors and equity returns.

Σκοπός της παρούσας μελέτης είναι να εξετάσουμε την σχέση που υπάρχει μεταξύ της χρηματιστηριακής αγοράς και αυτής των αποδόσεων των ομολόγων σε τέσσερις χώρες της Ευρωπαϊκής Ένωσης (Γερμανίας, Ιταλίας, Ισπανίας και Γαλλίας) για την περίοδο από τον Δεκέμβριο 1999 έως τον Δεκέμβριο του 2012. Προσπαθήσαμε να εξετάσουμε το κατά πόσο υπάρχουν συσχετίσεις μεταξύ των δύο περιουσιακών στοιχείων σε μεγάλο χρονικό διάστημα χρησιμοποιώντας πολυμεταβλητά μοντέλα. Τα δεδομένα που πήραμε είναι οι ημερήσιες αποδόσεις των 10ετών ομολόγων και τα κλεισίματα των χρηματιστηριακών αγορών των χωρών αυτών για κάθε μία ξεχωριστά. Ξεκινάμε την ερευνά μας χρησιμοποιώντας το μοντέλο του GARCH του Bollerslev (1990). Τέλος μέσω της συνολοκλήρωσης με την διαδικασία του Johansen test θα εξετάσουμε το κατά πόσο οι σειρές μας ολοκληρώνονται μακροχρόνια επηρεάζοντας η μία την άλλη καθώς και την μεταξύ τους εξάρτηση και την αιτιότητα των εν λόγω σχέσεων. Η εργασίας μας έχει ως στόχο να μας δείξει το κατά πόσο υπάρχει μακροχρόνια συσχέτιση μεταξύ των δύο αυτών αγορών, ώστε να βοηθά τους διαχειριστές και οικονομικούς αναλυτές να δημιουργούν το χαρτοφυλάκιο με το μικρότερο κίνδυνο και την μεγαλύτερη απόδοση. Τα αποτελέσματα μας δείχνουν μία μακροχρόνια συσχέτιση μεταξύ αυτών των δύο αγορών και ότι η μία αγορά επηρεάζει την άλλη σε βάθος χρόνου, οπότε είναι χρήσιμο σε ένα χαρτοφυλάκιο να υπάρχουν και τα δύο περιουσιακά στοιχεία. / The purpose of this study is to look at the relationship between stock market and bond market in four European Countries (Germany, France, Spain and Italy) for the period of December 1999 to December 2012. We attempt to examine whether the correlations between two classes of assets are time varying by using multivariate conditional volatility models. The data we are daily yields on 10-year bonds and the closures of the stock markets of these countries for each one individually. We start our investigation by applying GARCH model of Bollerslev (1990). Finally, through co integration with the process of Johansen test will look at whether our series completed long influencing each other and their mutual dependence and causality of these relations. Our paper aims to show us whether there is a long correlation between these two markets in order to help managers and financial analysts to create a portfolio with less risk and greater efficiency. Our results show a long-term correlation between these two markets and one market affects the other in the long run, so it is useful to have a portfolio of both assets.

The analysis of time series has long been the subject of interest in different fields. For decades time series were analysed with linear models, which have many advantages. Nevertheless, an issue which has been raised is whether there exist other models that can explain and forecast real data better than linear ones. In this thesis, new nonlinear time series models are suggested, which consist of a nonlinear conditional mean model, such as an ExpAR or an Extended ExpAR, and a nonlinear conditional variance model, such as an ARCH or a GARCH. Since new models are introduced, simulated series of the new models are presented, as it is important in order to see what characteristics real data which could be explained by them should have. In addition, the models are applied to various stationary and nonstationary economic and financial time series and are compared to the classic AR-ARCH and AR-GARCH models, in terms of fitting and forecasting. It is shown that, although it is difficult to beat the AR-ARCH and AR-GARCH models, the ExpAR and Extended ExpAR models and their special cases, combined with conditional heteroscedastic errors, can be useful tools in fitting, describing and forecasting nonlinear behaviour in financial and economic time series, and can provide some improvement in terms of both fitting and forecasting compared to the AR-ARCH and AR-GARCH models.

In the post Bretton Woods era, the volatile nature of exchange rates has been the focus of many researchers. Although some previous studies suggest that variations in an exchange rate has the potential to affect a country’s economic performance, LDC’s (Less Developed Countries’) have received less attention compared to industrialized or developed economies. In this thesis we analyse the nature of exchange rate behaviour in three LDCs: Ghana, Mozambique and Tanzania. These countries have gone through comparable policy engagements with the IMF, have followed similar floating exchange rate regimes since early 1990s and currently all adhere to the IMF convention of free current account convertibility and transfer (Ghana and Tanzania accepted Article VIII of IMF “Articles of Agreement” in 1994. Mozambique began floating in 1992 under the SAP reforms of IMF; Article IV consultation was completed in 2009 and acceptance of Article VIII seems imminent).The main content of the thesis can be summarised as follows.I. We examine whether exchange rate behaviour in these three countries are influenced by similar factors. In order to justify the applicability of a number of volatility modelling techniques, we also examine the data to find if they exhibit the empirical regularities found in other exchange rate/financial markets such as volatility clustering, non-linearity, non-normality and asymmetry. Our results suggest that exchange rate behaviour in these countries is generally influenced by similar factors. In particular, we find that the series exhibit the empirical regularities found in other exchange rate/financial markets, justifying the application of the ARCH methodology which we use to estimate the volatility of exchange rate in these countries. We however observed that the ARCH family of models does not always produce the best fit. For instance, volatility forecasts generated by an Exponentially Weighted Moving Average (EWMA) model based on the RiskMetricsTM estimation technique produces the best fit for the daily Ghanaian exchange rate series under consideration compared to volatility forecasts from our estimated ARCH family of models.II. We explore the causal relationship between exchange rate depreciation and uncertainty/volatility using the VAR toolkit. Our main motivation for this study is to analyse whether the changes in the levels of exchange rate as a result of appreciation or depreciation in an underlying currency changes the level of exchange rate uncertainty (volatility). Further, we also analyse the reverse causal relationship; whether increasing uncertainty feeds back into the exchange rate market. We find a bi-directional Granger causal relationship between the level of exchange rate and uncertainty in the foreign exchange markets. Despite adopting similar macro-policies since the mid 1980s and early 1990s, uncertainty in the Tanzanian exchange rate as a response to changes in the level of exchange rate takes a shorter length of time to dissipate. We attribute this to the macroeconomic policies undertaken by Tanzanian policymakers which have ensured price and currency stability.The reverse causality reflects the effectiveness of the Tanzanian macro-policies and the confidence in them; we observed that intervention reduces uncertainty in the Tanzanian exchange rate, whereas for Ghana and Mozambique, macro-policies intending to mitigate undesired exchange rate changes rather create further uncertainty in their exchange rate markets. For all three LDCs under consideration, we observed that effects of shocks to exchange rate from innovations in uncertainty for each country is fleeting III. We investigate the relationship between exchange rate volatility and economic performance (via trade) for each of these countries and some of their biggest trade partners. Exchange rate volatility resulting from a depreciating underlying currency of trade can potentially affect the economic performance of a country. Using a gravity model augmented with variables that are deemed to influence earnings from trade, we observe that earnings from trade are not significantly affected by exchange rate volatility. We conjecture that in periods of uncertainty, traders increase the volume of trade to compensate for the ill effects of currency volatility.

Abstract This thesis considers some linear and nonlinear time series models. In the linear case, the analysis of a large number of short time series generated by a first order autoregressive type model is considered. The conditional and exact maximum likelihood procedures are developed to estimate parameters. Simulation results are presented and compare the bias and the mean square errors of the parameter estimates. In Chapter 3, five important nonlinear models are considered and their time series properties are discussed. The estimating function approach for nonlinear models is developed in detail in Chapter 4 and examples are added to illustrate the theory. A simulation study is carried out to examine the finite sample behavior of these proposed estimates based on the estimating functions.