The first essay explores the dynamic behaviors of mortgage-backed stock returns and their volatility spillovers within the framework of time-varying symmetric, asymmetric and multivariate GARCH-family models. The focus of the chapter is on the dynamics of volatility of the U.S. real estate investment trusts (REITs) and volatility spillovers within the REITs subdivisions as well as between the REITs and the Fannie Mae (FNM) and theFreddie Mac (FRE) mortgage-backed stocks. We analyze risk-return linkages using the GARCH-in-mean (GARCH-M) model. The presence of asymmetric effects of "bad" news and "good" news on conditional financial volatilities is evaluated using the Threshold ARCH (TARCH) model and the exponential GARCH (EGARCH) model. Volatility spillovers and comovements within REITs subdivisions; REITs with FNM and FRE and other selected financial assets are examined using the multivariate GARCH (MGARCH) model. The second essay investigates factors behind the existence of time-varying conditional volatilities of mortgage-backed securities (MBS). This is done by analyzing the impacts of economic volatilities on mortgage-backed financial markets' performance. The relationship between conditional volatilities of the MBS and conditional volatilities of the key economic fundamentals in the housing sector and the macroeconomy are explored. The sensitivity of mortgage-backed stocks to the underlying time-series changes in economic fundamentals, and the extent to which economic volatilities explain the variation in mortgage-backed stocks' volatilities are investigated. Particularly, we examined whether changes in the REITs, FNM and FRE volatilities are linked to and driven by time-varying volatilities of the housing sector economic activity and set of key macroeconomic variables. Thus, the chapter analyzes the impacts of conditional economic volatilities on the conditional volatilities of the REITs, FRE and FNM stocks. The GARCH (p, q) process is used to find conditional volatility dynamics for the economic variables in the study. Then we employ multivariate GARCH (p, q) model to investigate the spillovers and comovements among the conditional economic fundamentals' volatilities and the conditional volatilities of the MBS. The third essay explores the impacts of foreign sector of the economy on the mortgage-backed financial markets and the housing sector. There is large surge of foreign capital flows to the U.S, particularly since late 1990s. The net foreign holdings of U.S. financial assets have become very significant in the U.S. Treasury notes and bonds. Foreign investors also hold a growing share of securities of the U.S. agencies and government sponsored enterprises (GSEs). Similarly, foreign direct investment in the U.S. real estate as well as real estate equities in the form of REITs has grown sharply. To this end, a multivariate vector autoregression (VAR) model is the main tool of analysis. Based on the VAR model, generalized impulse response functions and generalized variance decompositions are employed to evaluate the responses of mortgage interest rates and Treasury yields to the changes in net foreign ownership of U.S. Treasuries and agency bonds.

[pt] O expressivo fluxo de capital externo direcionado aos países emergentes ao longo dos anos 90 foi, e continua sendo, objeto de estudo de muitas pesquisas acadêmicas desenvolvidas na área de finanças internacionais. A maioria destes estudos busca investigar empiricamente, em que medida os fatores econômicos domésticos e externos constituem-se em determinantes significativos do fluxo internacional de capital. Semelhantemente, esse mesmo tipo de análise tem sido freqüentemente realizado sob a ótica do risco-país. A única diferença entre as duas análises é que de um lado o interesse é em torno dos determinantes do movimento de capital, do outro, o foco é sobre o movimento dos preços dos ativos e dos bônus soberanos. Dado que uma parte representativa dos fluxos de capital, ao longo da década de 90, esteve direcionada para o mercado de títulos da dívida pública dos países emergentes, motivada pelos mais diversos fatores, o presente trabalho tem como finalidade estudar essa dinâmica de preços com base em três modelos: i) o primeiro modelo analisa os determinantes econômicos internos do risco-país; ii) o segundo modelo tem o mesmo propósito do primeiro, sendo que a variável grau de aversão ao risco global, que serve de proxy para o componente externo do risco, é incluída no grupo de variáveis explicativas; iii) no último modelo, o enfoque é sobre a relação entre o risco-país específico e os fundamentos econômicos. Entendendo-se como risco específico, o risco-país menos o componente externo. No caso, o exercício em questão será aplicado para uma amostra de países emergentes e especificamente para o Brasil. / [en] The expressive flow of external capital directed to the emerging countries along 90's has been the subject of much academic research developed in the area of international finance. Most of these studies attempt to empirically investigate how much country-especific and global factors constitute significant determinants of capital flow. Similarly, this kind of analysis has frequently taken place under the view of country risk. The only difference between the two analyses is that on the one side the major interest is around the movements of the capital flow, whereas on the other one it is around the price determination of sovereign bonds. Given the fact that an important part of capital flow along the 90's was linked to external debt bonds of emerging market countries motivated by several reasons, this dissertation has the goal of studying the economic determinants of price variations based on three models: i) the first model analyzes the domestics determinants of country risk; ii) the second model has the same goal of the first one; however, the global risk aversion variable, which serves as proxy for the external risk component, is included to the group of explanatory variables; iii) the last model specifically focuses on the relationship between country risk and economic fundamentals. It is important that we understand as specific risk, the country risk deprived from the external components. Therefore, the following analysis will be applied to a sample of emerging markets and especifically to Brazil.

本論文的研究含括了墨西哥金融危機、亞洲金融危機與俄羅斯金融危機，主要是希望能歸納出90年代新興市場國家發生金融危機的普遍性原因。首先本論文先探討分析文獻上三大類型的金融危機理論，含括金融危機可預測與不可預測的原因。本論文試圖將可預測的經濟基本面歸為內部因素，不可預測的國際資金變動歸為外部因素。當一新興市場國家正在從事經濟結構改革或經濟轉型，而資本市場架構與金融機構監督機制與規範都尚未健全時，如果該國將金融市場自由化並允許大筆外資流入國內金融體系，則短期、投機性國際資金將會扭曲金融結構，造成經濟基本面的惡化。當不可預測的國際資金環境轉變時，一方面受到資金逆流衝擊著脆弱的金融部門，另一方面已經不健全的經濟基本面將嚴重惡化。此時政府採取的提高利率政策與賣出外匯捍衛本國釘住匯率制，卻會同時嚴重傷及本國銀行體系，促使金融機構面臨倒閉、陷入流動性不足的危機，結果投資者的信心仍然潰散而國內則造成金融體系的崩潰，終而引發全面性的金融危機。在建立了初步的研究假設後，我們以墨西哥金融危機與亞洲金融危機作為驗證。我們的研究案例中將包括危機發生的國家與當時未發生危機的國家作為對照組。在墨西哥危機，我們以智利作為對照組。在亞洲五國危機（印尼、韓國、馬來西亞、菲律賓與泰國），我們以中國大陸、新加坡、台灣作為對照組。在確立了我們的理論模型後，我們以個案研究的方式深入瞭解俄羅斯特殊的轉型經濟體質。俄羅斯激烈的經濟結構改革措施，使國家從金融封閉整合入世界資本市場。俄羅斯面臨著外資大量流入金融市場，然而總體經濟面卻在財政與貨幣政策相互矛盾的狀況下，實質經濟依舊無法發展起來，而金融體質嚴重惡化的情況。俄羅斯的案例也同樣證明本論文的論點，一個新興市場國家資本帳的開放必須有其先決條件且循序漸進的開放。一國能否將國際資金導入對於實質產業有所助益的直接投資，是新興市場國家所面臨的重要課題。不然，資本自由化先決條件的不足而快速地實施自由化，將只是加速金融危機的到來。 / This research has included the financial crises of Mexico, Asia and Russia, mainly expecting to conclude the general causes for the financial crises of emerging market economies in the 1990s. First, the paper reviews the three types of theoretical literature on financial crises, including predictable and unpredictable causes. This thesis tries to refer to the predictable economic fundamental part as internal factors, and the unpredictable, international capital mobility as external factors. When an emerging market country is engaged in economic structural reform or economic transition, and when the capital-market architecture and the system of financial-sector supervision and regulation are not sound, if the country liberalize financial market and allow large-scale foreign capital to inflow into domestic financial system, then short-term, speculative international capital will deteriorate financial structure and the economic fundamentals will become worse. When the environment of unpredictable international capital changes, on the one hand, the withdrawal of foreign capital impacts the vulnerable financial sector, and on the other hand, the unhealthy economic fundamentals will become seriously worse. At the same time, the government takes measures to raise interest rates and to sell foreign currency to defend the pegged exchange rate, which will seriously weaken the banking system, causing the bankrupt and the crisis of illiquidity. As a result, the confidence of investors will crash down and the native financial institutions will collapse, and this gives rise to the financial crisis in large scale. After the presumption of this research, we take Mexican and Asian financial crises for example. We try to give corresponding cases for contrast. In Mexican crisis, we take Chile as a contradistinctive group. In five Asian country crises (Indonesia, Korea, Malaysia, Philip, and Thailand), we take China, Singapore, and Taiwan as a contradistinctive group. After proving our model, we try to take Russia as a case study because of the special nature of economic transformation of Russia. The policy of drastic structural economic reform has made Russia step from the financial isolation to the integration into World’s capital market. Russia faces the large-scale foreign capital inflow in the financial market, while the real economy can not recover and domestic financial institutions become structurally weak because of the contradiction of the fiscal and monetary policy. The case of Russia also shows us the argument of this thesis--capital-account liberalization of an emerging market must depend on its prerequisites and steadiness of liberalization. When a nation can make proper use of the capital, conducting foreign investment to real economy is the important challenge of an emerging market. Otherwise, the capital-account liberalization will speed up the coming of financial crisis.