L'ampleur de la Grande Récession a suscité un regain d'intérêt pour l'analyse conjoncturelle, plus particulièrement du cycle économique. Notre thèse participe de ce renouveau d'attention pour l'étude des fluctuations économiques.Après une présentation générale des modèles à changements de régime dans le chapitre 1, le chapitre suivant propose une chronologie du cycle des affaires de l'économie française sur la période 1970-2009. Trois méthodes de datation sont utilisées à cette fin : la règle des deux trimestres consécutifs de croissance négative, l'approche non paramétrique de Bry et Boschan (1971) et le modèle markovien à changements de régime de Hamilton (1989). Les résultats montrent que l'existence de ruptures structurelles peut empêcher ce dernier modèle d'identifier correctement les points de retournement cycliques. Cependant, quandces ruptures sont prises en considération, le calendrier des récessions françaises obtenu à l'aide du modèle d'Hamilton coïncide largement avec celui obtenu par les deux autres méthodes. Le chapitre 3 développe une analyse de la non-linéarité dans le modèle à changements de régime en utilisant un ensemble de tests non-standards. Une étude par simulation Monte Carlo révèle qu'un test récemment proposé par Carrasco, Hu et Ploberger (2013) présente une faible puissance pour des processus générateurs des données empiriquement pertinents et ce, lorsqu'on tient compte de l'autocorrélation sous l'hypothèse nulle. En revanche, untest "bootstrap" paramétrique basé sur le rapport des vraisemblances a, pour sa part une puissance plus élevée, ce qui traduit l'existence probable de non-linéarités significatives dans le PIB réel trimestriel de la France et des Etats-Unis. Quand il s'agit de tester un changement de régime en moyenne ou en constante, il est important de tenir compte de l'autocorrélation sous l'hypothèse nulle de linéarité. En effet, dans le cas contraire, un rejet de la linéarité pourrait simplement refléter une mauvaise spécification de la persistance des données, plutôt que d'une non-linéarité inhérente.Le chapitre 4 examine une question importante : la considération de ruptures structurelles dans les séries améliore-t-elle la performance prédictive du modèle markovien relativement à son homologue linéaire ? La démarche adoptée pour y répondre consiste à combiner les prévisions obtenues pour différentes périodes d'estimation. Voici le principal résultat dû à l'application de cette démarche : la prise en compte des données provenant des intervalles de temps précédant les ruptures structurelles et la "Grande Modération" améliore les prévisions basées sur des données tirées exclusivement de ces épisodes. De la sorte, les modèles à changements de régime s'avèrent capables de prédire la probabilité d'événements tels que la Grande Récession, avec plus de précision que ses homologues linéaires.Les conclusions générales synthétisent les principaux acquis de la thèse et évoqueplusieurs perspectives de recherche future. / The severity of the Great Recession has renewed interest in the analysis of business cycles. Our thesis pertains to this revival of attention for the study of cyclical fluctuations. After reviewing the regime-switching models in Chapter one, the following chapter suggests a chronology of the classical business cycle in French economy for the 1970-2009 period. To that end, three dating methodologies are used: the rule of thumb of two consecutive quarters of negative growth, the non-parametric approach of Bry and Boschan (1971), and the Markov-switching approach of Hamilton (1989). The results show that,omitted structural breaks may hinder the Markov-switching approach to capture business-cycle fluctuations. However, when such breaks are allowed for, the timing of the French recessions provided by the Markov-switching model closely matches those derived by the rule-based approaches.Chapter 3 performs a nonlinearity analysis inMarkov-switching modelling using a set of non-standard tests. Monte Carlo analysis reveals that a recently test proposed by Carrasco, Hu, and Ploberger (2013) for Markov switching has low power for empirically-relevant data generating processes when allowing for serial correlation under the null. By contrast, a parametric bootstrap likelihood ratio (LR) test of Markov switching has higher power in the same setting, providing stronger support for nonlinearity in quarterly French and U.S. real GDP. When testing for Markov switching in mean or intercept of an autoregressive process, it is important to allow for serial correlation under the null hypothesis of linearity.Otherwise, a rejection of linearity could merely reflect misspecification of the persistence properties of the data, rather than any inherent nonlinearity.Chapter 4 examines whether controlling for structural breaks improves the forecasting performance of the Markov-switching models, as compared to their linear counterparts.The approach considered to answer this issue is to combined forecasts across different estimation windows. The outcome of applying such an approach shows that, including data from periods preceding structural breaks and particularly the "Great Moderation" improves upon forecasts based on data drawn exclusively from these episodes. Accordingly, Markov-switching models forecast the probability of events such as the Great Recession more accurately than their linear counterparts.The general conclusions summarize the main results of the thesis and, suggest several directions for future research.

Suite à la crise financière de 2007, les outils de politique monétaire conventionnelle se sont avérés insuffisants pour stabiliser l'économie et empêcher la diffusion de la crise financière. Les banques centrales ont de fait mis en place des politiques monétaires non conventionnelles. L’objectif de cette thèse est de participer à la compréhension théorique et empirique des politique monétaires non conventionnelles en concentrant nos efforts de modélisation sur la nature non-linéaire de la crise financière. Les deux premiers chapitres de cette thèse développent des modèles DSGE incorporant des contraintes de crédit occasionnellement saturées de manière à capturer la nature transitoire des phénomènes de crise. Dans le premier chapitre - obtenu dans un cadre d'économie fermée - un résultat notable est que les politiques d’assouplissement quantitatif diminuent bien l’amplitude de la crise, mais augmentent sa durée. Dans le deuxième chapitre, lorsque l’on implémente des programmes d’assouplissement quantitatif au niveau d'une union monétaire hétérogène constituée de deux pays, se pose le problème des hétérogénéités entre les pays membres de cette union. Nous trouvons qu’il est toujours plus intéressant pour la banque centrale de concentrer ses achats de titres dans les pays de l’union les plus touchés par la crise financière. De plus, un niveau intermédiaire d’intégration financière permet de minimiser les conséquences de la crise au niveau de l’union monétaire dans son ensemble. Dans le troisième chapitre, nous utilisons un modèle Markov-Switching VAR Bayésien pour comparer l’efficacité des politiques d’assouplissement quantitatif en période de crise et en période normale. Alors que les programmes d’assouplissement quantitatif sont particulièrement efficaces en période de crise, nous ne trouvons aucun effet significatif de ces programmes sur les variables macroéconomiques lorsque l'économie retourne à son état initial. / Following the 2007 financial crisis, conventional monetary policy tools prooved insufficient to stabilize the macroeconomy and to avoid a financial disruption. As a consequence, central banks relied more heavily on unconventional monetary policy tools. This thesis aims at contributing to the understanding of unconventional monetary policy tools, focusing on the inherently non-linear nature of financial crises. In the first two chapters, we use DSGE models with occasionally binding credit constraints to account for the transitory nature of financial disruption events. In chapter one, in the case of a closed economy, we find that quantitative easing decreases the magnitude of the crisis but increases its duration. Still, when looking for intertemporal effects of quantitative easing programs, it appears that they are always welfare improving. In chapter two, when implementing quantitative easing on a two country monetary union, comes the question of how to deal with heterogeneities between members. We find that it is always better to implement nationaly tailored quantitative easing programs. Finally, an intermediate degree of financial integration proves optimal to dampen the macroeconomic consequences of the financial crisis on the overall monetary union. In the third chapter, we use a Markov-Switching Bayesian VAR model to compare the efficiency of quantitative easing in normal times versus financial crisis times. While quantitative easing programs are highly efficient during financial crisis times, we find no significant effect of these programs when the economy goes back to normal times.

Problem statement: Accurate and reliable forecasts of trends in the housing market can be useful information for market participants as well as policy makers. This information may be useful to minimize risk related to market uncertainty. Since the burst of the housing bubble in the early 1990s the price level of single-family houses has risen sharply in Sweden. The Swedish housing market has experienced an unusually long period of high growth rates in transaction prices which has opened up for discussions about the risk of another housing bubble. Business and property cycles have shown to contain asymmetries, which linear models are unable to pick up and therefore inappropriate to analyze cycles. Approach: Therefore, this study uses non-linear models which are able to pick up the asymmetries. The estimated models are variations of the Markov-switching regression model, i.e. the Markov-switching autoregressive (MS-AR) model and the Markov-switching dynamic regression (MS-DR) model. Results: Our ndings show that the MS-AR(4) model allowing for varying variance across regimes estimated using the growth rate of FASTPI produce superior forecasts over other MSAR models as well as variations of the MS-DR model. The average expected duration to remain in a positive growth regime is between 6.3 and 7.3 years and the average expected duration to remain in a negative growth regime is between 1.2 to 2.5 years. Conclusion: The next regime shift in the Swedish housing market is projected to occur between 2018 and 2019, counting the contraction period in 2012 as the most recent negative regime. Our ndings support other studies ndings which indicate that the longer the market has remained in one state, the greater is the risk for a regime shift. / Problemformulering: Noggranna och tillforlitliga prognoser om utvecklingen pa bostadsmarknaden kan vara anvandbar information for marknadsaktorer samt beslutsfattare. Denna information kan vara anvandbar for att minimera risken relaterad till osakerheten pa marknaden. Sen bostadsbubblan sprack i borjan av 1990-talet har prisnivan for smahus okat kraftigt i Sverige. Den svenska bostadsmarknaden har upplevt en ovanligt lang period av hog tillvaxt i transaktionspriser som har oppnat upp for diskussioner om risken for en ny bostadsbubbla. Konjunkturoch fastighetscykler har visat sig innehalla asymmetrier som linjara modeller inte kan uppfanga och darfor visat sig vara olampliga for att analysera cykler. Tillvagagangssatt: Darfor anvander den har studien icke-linjara modeller som kan uppfanga dessa asymmetrier. De skattade modellerna ar variationer av Hamiltons Markov-switchingmodell, dvs. en autoregressiv Markov-switchingmodell (MS-AR) och en dynamisk Markov-switchingmodell (MS-DR). Resultat: Resultatet visar att MS-AR(4)-modellen som tar hansyn till varierande varians over regimerna estimerad med tillvaxten av FASTPI producerar overlagsna prognoser jamfort med andra MS-AR-modeller samt variationer av MS-DR-modellen. Den genomsnittliga forvantade varaktigheten att benna sig i en positiv regim ar mellan 6,3 och 7,3 ar och den  genomsnittliga forvantade varaktigheten att benna sig i en negativ regim ar mellan 1,2 till 2,5 ar. Slutsats: Nasta regimskifte pa den svenska bostadsmarknaden beraknas ske mellan 2018 och 2019, antaget att nedgangen under 2012 ar den senaste negativa regimen. Resultatet stodjer tidigare studier, som tyder pa att ju langre marknaden har varit i ett tillstand, desto storre ar risken for ett regimskifte.

Understanding financial asset return correlation is a key facet in asset allocation and investor’s portfolio optimization strategy. For the last decades, several studies have investigated this relationship between stock and bond returns. But, fewer studies have dealt with multi-asset return dynamics. While initial literature attempted to understand the fundamental pattern of comovements, later studies model the economic state variables influencing such time-varying comovements of primarily stock and bond returns. Research widely acknowledges that return distributions of financial assets are non-normal. When the joint distributions of the asset returns follow a non-elliptical structure, linear correlation fails to provide sufficient information of their dependence structure. In particular two issues arise from this existing empirical evidence. The first is to propose a more reliable alternative density specification for a higher-dimensional case. The second is to formulate a measure of the variables’ dependence structure which is more instructive than linear correlation. In this work I use a time-varying conditional multivariate elliptical and non-elliptical copula to examine the return comovements of three different asset classes: financial assets, commodities and real estate in the US market. I establish the following stylized facts about asset return comovements. First, the static measures of asset return comovements overestimate the asset return comovements in the economic expansion phase, while underestimating it in the periods of economic contraction. Second, Student t-copulas outperform both elliptical and non-elliptical copula models, thus confirming the ii dominance of Student t-distribution. Third, findings show a significant increase in asset return comovements post August 2007 subprime crisis ... [cont.].

A evolução da economia brasileira durante os primeiros anos do século XXI é examinada através de um modelo microfundamentado de uma pequena economia aberta, permitindo mudanças no comportamento do Banco Central do Brasil, no parâmetro de rigidez nominal e na volatilidade dos choques estruturais. Mesmo os resultados não sendo conclusivos a respeito da presença de mudanças de regime durante o período analisado, encontramos evidências de troca de regime no âmbito da política monetária, passando em 2003 de um regime Dove para um regime Hawk, assim como evidências de choques externos mais voláteis durante períodos de incerteza. Na sequência, deixamos de lado a estimação empírica e derivamos regras de política monetária ótima para o caso brasileiro. É possível encontrar uma regra ótima capaz de estabilizar a inflação, o produto e a taxa de câmbio, mantendo uma taxa de juros estável. Por fim, o modelo trás uma discussão interessante sobre a dinâmica de determinadas variáveis macroeconômicas: uma moeda mais estável implica em uma taxa de juros mais volátil, e vice versa; um maior controle sobre a taxa de juros e/ou sobre a taxa de câmbio parece gerar uma maior instabilidade do produto e da inflação. / The evolution of the Brazilian economy during the first years of this century is examined through the lens of a micro-founded small open economy model that allows for changes in the behaviour of the Central Bank of Brazil, in the nominal price rigidity and in the volatility of structural shocks. Although the results are not conclusive about the presence of regime changes during the analysed sample, we find evidences in favour of shifts in the monetary policy stance, moving from a Dove to a Hawk regime in 2003, as well as evidences of more volatile external shocks during uncertainty periods. We further move away from the empirical estimation and derive optimal monetary policy rules for Brazil. It is possible to find an optimal rule that is successful in stabilizing inflation, output and exchange rates, whilst keeping interest rates stable. Finally, the model offers interesting insights about the standard deviation dynamics of macroeconomic variables: a more stable currency implies a more volatile interest rate and vice versa, and a higher control over interest rates and/or exchange rates seem to produce output and inflation instability.

在觀察匯率市場是否具有效率性時，大部分文獻透過檢定「遠期匯率是否為未來即期匯率的不偏估計值」來驗證，然而實證結果多不支持。探究原因後，部分學者於是提出，可能是在效率市場的假設上出了問題。原效率市場假設理性預期與風險中立，可是在現實生活中，人們的行為大多顯現風險趨避的特質，學者因而推論「風險溢酬的存在」或許正是造成遠期匯率偏誤的原因。 　Lucas（1982）在跨期資本資產訂價理論推導中證明出，風險溢酬具有因時而異的性質。Domowitz and Hakkio（1985）對該理論做進一步設定後，得到風險溢酬為兩國間貨幣政策波動差異的函數，因而改良風險溢酬模型為受到匯率預測精確性影響，並以ARCH模型估計。 　本文承續Domowitz and Hakkio（1985）的理論設定，以市場風險解釋風險溢酬，同時引進馬可夫變換模型，用以捕捉因時而異的風險溢酬，並且將其與ARCH-M模型所估計出的風險溢酬加以比較，期望能找出一個對風險溢酬解釋力較佳的模型。

According to Kim and Nelson (1999) and McConnell and Perez-Quiros (2000), the timing of the Great Moderation occurred in U.S. at 1984Q1. Summers (2005) found out several reasons and different timings of the Great Moderation in the G-7 countries and Australia. During the past fifty years, there was a significantly sharp decline in the volatility of the real growth rate in Australia. Between 1968 and 1982, the standard deviation of the real growth rate was 1.416%¡Fhowever, between 1983 and 1996, the standard deviation of the real growth rate drastically reduced to 0.917%. Based on this obvious situation described above, we successively build up a Markov-Switching Model and Time-Varying Structural Autoregressive Model to investigate the structural break and the sources of the Great Moderation in Australia. The findings turn out that improved monetary policy and the decreased oil shock can account for the explanation of the moderation with the break date of 1984Q1.

Deep penetration of personal computers, data communication networks, and the Internet has created a massive platform for data collection, dissemination, storage, and retrieval. Large amounts of textual data are now available at a very low cost. Valuable information, such as consumer preferences, new product developments, trends, and opportunities, can be found in this large collection of textual data. Growing worldwide competition, new technology development, and the Internet contribute to an increasingly turbulent business environment. Conducting surveillance on this growing collection of textual data could help a business avoid surprises, identify threats and opportunities, and gain competitive advantages.Current text mining approaches, nonetheless, provide limited support for conducting surveillance using textual data. In this dissertation, I develop novel text quantification approaches to identify useful information in textual data, effective anomaly detection approaches to monitor time series data aggregated based on the text quantification approaches, and empirical evaluation approaches that verify the effectiveness of text mining approaches using external numerical data sources.In Chapter 2, I present free-text chief complaint classification studies that aim to classify incoming emergency department free-text chief complaints into syndromic categories, a higher level of representation that facilitates syndromic surveillance. Chapter 3 presents a novel detection algorithm based on Markov switching with jumps models. This surveillance model aims at detecting different types of disease outbreaks based on the time series generated from the chief complaint classification system.In Chapters 4 and 5, I studied the surveillance issue under the context of business decision making. Chapter 4 presents a novel text-based risk recognition design framework that can be used to monitor the changing business environment. Chapter 5 presents an empirical evaluation study that looks at the interaction between news sentiment and numerical accounting earnings information. Chapter 6 concludes this dissertation by highlighting major research contributions and the relevance to MIS research.

Made available in DSpace on 2010-04-20T20:58:04Z (GMT). No. of bitstreams: 4 2006 - Wagner_Oliveira_ Monteiro_02_10_2008.pdf.jpg: 17677 bytes, checksum: 012a0852290fa51f423a5a8ec7534ea5 (MD5) 2006 - Wagner_Oliveira_ Monteiro_02_10_2008.pdf: 450170 bytes, checksum: ea37b352c4028dd1c20da87d3f3badf2 (MD5) 2006 - Wagner_Oliveira_ Monteiro_02_10_2008.pdf.txt: 55718 bytes, checksum: 579a00e43cb84159205c5d87713ad640 (MD5) license.txt: 4884 bytes, checksum: de2d265ed2868529ac27feb118588da8 (MD5) Previous issue date: 2008-10-02T00:00:00Z / This dissertation proposes a bivariate markov switching dynamic conditional correlation model for estimating the optimal hedge ratio between spot and futures contracts. It considers the cointegration between series and allows to capture the leverage efect in return equation. The model is applied using daily data of future and spot prices of Bovespa Index and R$/US$ exchange rate. The results in terms of variance reduction and utility show that the bivariate markov switching model outperforms the strategies based ordinary least squares and error correction models.

O objetivo geral deste trabalho é testar se a informação contida em dados de microestrutura de mercado contribui para uma melhor explicação do comportamento dos preços dos títulos negociados na BMF&BOVESPA. O primeiro ensaio testa a hipótese de que o volume gera informação sobre o movimento dos preços quando os traders possuem informações diferentes sobre o comportamento dessa variável. Em uma abordagem de microestrutura de mercado, os traders não informados observam apenas as estatísticas diárias de preço e volume, e os traders informados conhecem o comportamento dos preços e volumes tick-by-tick. Aplicando o algoritmo do tick-test, os traders informados observam o volume de ordens de compras e vendas diárias. O teste empírico é feito utilizando modelos markov-switching, e os resultados revelam que os traders informados apresentam estimações melhores. No segundo ensaio, buscam-se evidências de que traders informados preferem atuar no mercado de opções pela possibilidade de alavancar seus ganhos. O objetivo é identificar se o volume no mercado de opções revela o comportamento dos preços das ações. Os resultados obtidos, utilizando opções de PETR4 e VALE5, corroboram a hipótese que o volume no mercado de opções não transmite informação sobre o comportamento do preço da ação. O terceiro ensaio utiliza dados tick-by-tick para estimar a probabilidade de informação privilegiada (PIN) para ações do IBRX. O PIN é uma proxy para informação privada e é incorporado ao método de Fama e French (1993) para separar os portfolios e explicar seus retornos. A combinação do PIN com as variáveis valor de mercado e índice book-to-market possibilita, para alguns portfolios, aumento do retorno e diminuição do risco. A significância das variáveis na explicação dos retornos é testada utilizando modelos de markov-switching. Os resultados permitem concluir que o PIN é um fator importante na explicação dos retornos dos portfolios. / The general objective is to test whether the information contained in data of microstructure market contributes to a better explanation of the behavior of stock prices negotiated in BMF&BOVESPA. The first essay tests the hypothesis that volume generates information on price movements when traders have different information on the behavior of this variable. In a model of market microstructure, non-informed traders observe only the daily statistics on price and volume, while informed traders know the behavior of tick-by-tick prices and volumes. Applying the tick-test algorithm, informed traders observe the volume of buy orders and daily sales. The empirical test is carried out by using markov-switching models, and the results reveal that informed traders show better estimates. In the second essay, we look for evidence that informed traders prefer to trade in options market due to its possibility of leveraging their revenue. The objective is to identify whether volume in the options market is revealing of the behavior of stock prices. The results obtained using options from PETR4 and VALE5 concur to the hypothesis that volume in options market does not transmit information on the behavior of stock price. The third essay makes use of tick-by-tick data to estimate the probability of privileged information (PIN) to IBRX stocks. PIN is a proxy to privilege information and is incorporated to Fama French (1993) method to separate portfolios and explain their returns. The combination of PIN with variables of market value and book-to-market index allows some portfolios to increase return and diminish risk. The significance of these variables in explaining returns is tested using markov-switching methods. The results demonstrate that PIN is an important factor in explaining portfolio returns.