Il y a un consensus que la sévérité des chocs sur les marchés pétroliers tend à diminuer, ainsi que la dépendance des économies développées vis-à-vis de ces chocs. Les pays développés sont généralement les importateurs d'énergie et l'effet des chocs pétroliers sur les pays exportateurs de pétrole peut être différent, surtout s’il s’agit des pays dont la grande partie de l’exportation est le pétrole ou les produits pétroliers. En outre, l'orientation sur l'exportation des matières premières peut modifier la performance relative des modèles économétriques qui sont généralement utilisés pour les prévisions. La thèse étudie et développe des modèles de l'analyse structurelle et de la prévision à court terme d'une économie exportatrice de pétrole où les données russes sont utilisées pour toutes les applications empiriques. Le premier chapitre est consacré à la construction d'un modèle DSGE pour un pays exportateur de matières premières. Le modèle DSGE est estimé par des méthodes bayésiennes. Nous constatons qu'en dépit de l'impact important sur le PIB des chocs pétroliers, les cycles économiques en Russie sont essentiellement d'origine intérieure. Le deuxième chapitre examine comment les méthodes bayésiennes peuvent être appliquées aux prévisions à l'aide d'un modèle BVAR. Le troisième chapitre applique ces techniques et compare la performance d'un groupe de modèles non structurels (univariés et multivariés) pour prévoir un ensemble d'indicateurs macroéconomiques russes. Dans le quatrième chapitre, les prévisions se sont concentrées sur les modèles structurels multivariés (DSGE) et non structurels (BVAR). Le cinquième chapitre quantifie l'effet de différents types de chocs pétroliers sur plusieurs variables macroéconomiques russes. / It is generally agreed that the severity of oil markets shocks tends to decrease as does dependence of developed economies on those shocks. Developed countries are generally energy importers, and the effect of oil market shocks on oil-exporting countries may be different, especially if energy represents a large percentage of the country’s exports. In addition, the focus on commodity exports may change the relative forecasting performance of econometric models that are generally used for forecasting. This thesis studies and develops models for structural analysis and short-term forecasting of an oil-exporting economy using Russian data for all empirical applications. The first chapter is devoted to a construction of a DSGE model for a country with commodity exports. The DSGE model is estimated by Bayesian methods We find that despite a strong impact of commodity export shocks on GDP, the business cycles in Russia are mostly domestically based.. The second chapter discusses how the Bayesian methods may be applied for forecasting with a BVAR model. The third chapter applies these techniques and compares the performance of a group of non-structural models – univariate and multivariate – for forecasting a set of Russian macroeconomic indicators. In the fourth chapter, the forecasting focuses on multivariate structural (DSGE) and non-structural BVAR models. The fifth chapter quantifies the effect of different types of oil market shocks on several Russian macroeconomic variables.

As pesquisas em finanças tem focado nos últimos anos a modelagem da variância devido à dificuldade de se obter boa previsão dos retornos na média, negligenciando o uso deste último, quanto informação necessária, no estabelecimento de estratégias de alocação de ativos em carteiras de renda variável. Seguindo DeMiguel, Nogales e Uppal (2014), esse trabalho adota o VAR como meio de se obter previsões dos retornos um passo à frente, e então, usá-las na alocação dos ativos. Para a inserção do VAR em finanças, foi permitido que os parâmetros variassem no tempo, o que conseguiu captar com sucesso a dinâmica volátil do mercado de ações. Ainda foram incorporadas técnicas bayesianas de estimação, a fim de driblar a sobreparametrização e obter estimações mais suavizadas, evitando deste modo que a variância das carteiras fossem muito altas. O método teve sucesso na aplicação e mostrou que o uso da previsão um passo à frente para os retornos pode ser usada como uma boa estratégia, expressa nos altos índices de Sharpe encontrados. / Research in finance has focused in recent years in variance modelling due to the difficulty of obtaining good forecasts of mean returns, neglecting the use of this latter on the establishment of asset allocation strategies in the equity portfolios. Following DeMiguel, Nogales e Uppal (2014), this work adopts the VAR as a way of obtaining forecasts of returns one step ahead and to get use of them in the allocation of assets. For insertion of the VAR in finance, the parameters was allowed to vary over time, which successfully captured the volatile dynamics of the stock market. Bayesian estimation techniques was incorporated in order to surmount overparameterization and to get more smoothed estimates, thereby preventing the variance of the portfolio to be very high. The method was successful in this application and showed that the making of one step ahead prediction of returns can be used as a good strategy, which can be expressed in high levels of Sharpe.

As pesquisas em finanças tem focado nos últimos anos a modelagem da variância devido à dificuldade de se obter boa previsão dos retornos na média, negligenciando o uso deste último, quanto informação necessária, no estabelecimento de estratégias de alocação de ativos em carteiras de renda variável. Seguindo DeMiguel, Nogales e Uppal (2014), esse trabalho adota o VAR como meio de se obter previsões dos retornos um passo à frente, e então, usá-las na alocação dos ativos. Para a inserção do VAR em finanças, foi permitido que os parâmetros variassem no tempo, o que conseguiu captar com sucesso a dinâmica volátil do mercado de ações. Ainda foram incorporadas técnicas bayesianas de estimação, a fim de driblar a sobreparametrização e obter estimações mais suavizadas, evitando deste modo que a variância das carteiras fossem muito altas. O método teve sucesso na aplicação e mostrou que o uso da previsão um passo à frente para os retornos pode ser usada como uma boa estratégia, expressa nos altos índices de Sharpe encontrados. / Research in finance has focused in recent years in variance modelling due to the difficulty of obtaining good forecasts of mean returns, neglecting the use of this latter on the establishment of asset allocation strategies in the equity portfolios. Following DeMiguel, Nogales e Uppal (2014), this work adopts the VAR as a way of obtaining forecasts of returns one step ahead and to get use of them in the allocation of assets. For insertion of the VAR in finance, the parameters was allowed to vary over time, which successfully captured the volatile dynamics of the stock market. Bayesian estimation techniques was incorporated in order to surmount overparameterization and to get more smoothed estimates, thereby preventing the variance of the portfolio to be very high. The method was successful in this application and showed that the making of one step ahead prediction of returns can be used as a good strategy, which can be expressed in high levels of Sharpe.

As pesquisas em finanças tem focado nos últimos anos a modelagem da variância devido à dificuldade de se obter boa previsão dos retornos na média, negligenciando o uso deste último, quanto informação necessária, no estabelecimento de estratégias de alocação de ativos em carteiras de renda variável. Seguindo DeMiguel, Nogales e Uppal (2014), esse trabalho adota o VAR como meio de se obter previsões dos retornos um passo à frente, e então, usá-las na alocação dos ativos. Para a inserção do VAR em finanças, foi permitido que os parâmetros variassem no tempo, o que conseguiu captar com sucesso a dinâmica volátil do mercado de ações. Ainda foram incorporadas técnicas bayesianas de estimação, a fim de driblar a sobreparametrização e obter estimações mais suavizadas, evitando deste modo que a variância das carteiras fossem muito altas. O método teve sucesso na aplicação e mostrou que o uso da previsão um passo à frente para os retornos pode ser usada como uma boa estratégia, expressa nos altos índices de Sharpe encontrados. / Research in finance has focused in recent years in variance modelling due to the difficulty of obtaining good forecasts of mean returns, neglecting the use of this latter on the establishment of asset allocation strategies in the equity portfolios. Following DeMiguel, Nogales e Uppal (2014), this work adopts the VAR as a way of obtaining forecasts of returns one step ahead and to get use of them in the allocation of assets. For insertion of the VAR in finance, the parameters was allowed to vary over time, which successfully captured the volatile dynamics of the stock market. Bayesian estimation techniques was incorporated in order to surmount overparameterization and to get more smoothed estimates, thereby preventing the variance of the portfolio to be very high. The method was successful in this application and showed that the making of one step ahead prediction of returns can be used as a good strategy, which can be expressed in high levels of Sharpe.

Cette thèse s’articule autour de trois essais consacrés à l’analyse de la surliquidité bancaire dans les pays de la CEMAC. Le premier essai a cherché à identifier les déterminants de la surliquidité des pays membres de la CEMAC sur la période de 1985 à 2002. L’estimation GMM utilisé a permis de montrer que la surliquidité en Afrique Centrale dérive à la fois du comportement de précaution des banques commerciales et des facteurs exogènes. La grande prudence des banques peut s’expliquer par l’expérience de la crise financière des années 1980, la restructuration du système bancaire, l’instabilité des dépôts et un contexte économique très risqué. L’embellie du cours du pétrole alimente les réserves excédentaires due à la faible capacité d’absorption des pays de la zone. Dans le deuxième essai, il a été question d’identifier les canaux de transmission les plus opérationnels en Afrique Centrale. La modélisation VAR a permis de montrer que le taux d’intérêt est le canal le plus faible. C’est précisément la carence d’un marché financier qui ne permet pas d’assurer le rôle de recyclage de la liquidité bancaire et de la transmission de la politique monétaire. Dans le troisième essai, a été élaboré un modèle de prévision d’inflation dans un des pays membres de la CEMAC à savoir le Tchad. Les modèles BVAR se sont révélés en Afrique Centrale être plus efficaces que les outils traditionnels (AR, ARIMA, VAR).L’analyse de sensibilité entreprise par l’approche bayesienne indique que la surliquidité exercerait des tensions inflationnistes dans la zone. / This thesis is based on three essays focused on analysis of the excess liquidity in the CEMAC countries. The first one identifies the determinants of excess liquidity in developing countries by studying the case of CEMAC member countries from 1985 to 2002. The GMM estimator used has shown that the excess liquidity in Central Africa derived from both the prudence of commercial banks and exogenous factors. The precaution of Commercial Banks can be explained by the financial crisis in 1980s, the restructuring of the banking system, the instability of deposits and a very risky economic environment. The increasing in oil prices fueling excess reserves due to the low absorption capacity of countries in the region. In the second essay, it was discussed the transmission channels in Central Africa. The VAR model used has shown that the interest rate channel is the lowest. This is exactly the lack of a financial market that does not allow ensuring the function of the recycling of liquidity and the transmission of mone ary policy. At the last essay, it was developed a forecasting model of inflation in Chad which is member of CEMAC countries. BVAR models have been shown in Chad to be more effective than traditional tools (AR, ARIMA and VAR). A sensitivity analysis undertaken by the Bayesian approach indicates that the excess liquidity would exert inflationary pressures.

Exports are one of the key aggregates in the Argentina’s economy, both because to its links with thedomestic demand and by its influence on the behaviour of the trade balance and current account.Have adequate forecasts for this variable is useful to design policies to keep surpluses in the externalsector and prevent recurring crises seen in the past. In this work, we considered some modelsfor forecasting the performance of this aggregate, which could be an alternative to the estimationof structural econometric models. For this purpose, we used two approaches: the first is based instandard and Bayesian VARs (Minnesota prior, Gibbs sampler, partial BVAR and BVAR-Kalman). Thelatter combines the evidence in the data with any prior information that may also be available. Thesecond approach considers the FAVAR (Factor-augmented VAR) models, which combines the standardVAR with factor analysis. Finally, we evaluated the forecasting ability of different models. / Las exportaciones representan uno de los agregados más importantes de la economía argentina,tanto por su vinculación con la demanda doméstica como por su influencia en el comportamientode la balanza comercial y de la cuenta corriente. Disponer de adecuados pronósticos deesta variable resulta útil a fin de diseñar políticas que permitan mantener superávit en el sectorexterno y evitar las recurrentes crisis observadas en el pasado. En este trabajo, se consideran algunosmodelos destinados a la realización de pronósticos de dicho agregado, los cuales podrían seruna alternativa a la estimación de sistemas econométricos estructurales. A tal efecto, se utilizandos propuestas: la primera se basa en modelos de VAR sin restricciones y Bayesianos (‘Minnesota’prior, ‘Gibbs sampler’, parcial BVAR y BVAR-Kalman). Estos últimos consideran supuestos a priori(‘prior’) e información histórica de las series de tiempo empleadas. La segunda propuesta descansaen modelos FAVAR (Factor-aumentado VAR), que combinan los VAR con el análisis de factores.Finalmente, se evalúa la capacidad de pronóstico de los distintos modelos.

Cette thèse s'articule autour de trois essais consacrés à l'analyse de la surliquidité bancaire dans les pays de la CEMAC. Le premier essai a cherché à identifier les déterminants de la surliquidité des pays membres de la CEMAC sur la période de 1985 à 2002. L'estimation GMM utilisé a permis de montrer que la surliquidité en Afrique Centrale dérive à la fois du comportement de précaution des banques commerciales et des facteurs exogènes. La grande prudence des banques peut s'expliquer par l'expérience de la crise financière des années 1980, la restructuration du système bancaire, l'instabilité des dépôts et un contexte économique très risqué. L'embellie du cours du pétrole alimente les réserves excédentaires due à la faible capacité d'absorption des pays de la zone. Dans le deuxième essai, il a été question d'identifier les canaux de transmission les plus opérationnels en Afrique Centrale. La modélisation VAR a permis de montrer que le taux d'intérêt est le canal le plus faible. C'est précisément la carence d'un marché financier qui ne permet pas d'assurer le rôle de recyclage de la liquidité bancaire et de la transmission de la politique monétaire. Dans le troisième essai, a été élaboré un modèle de prévision d'inflation dans un des pays membres de la CEMAC à savoir le Tchad. Les modèles BVAR se sont révélés en Afrique Centrale être plus efficaces que les outils traditionnels (AR, ARIMA, VAR). L'analyse de sensibilité entreprise par l'approche bayesienne indique que la surliquidité exercerait des tensions inflationnistes dans la zone.

This paper proposes a large Bayesian Vector Autoregressive (BVAR) model with common stochastic volatility to forecast global equity indices. Using a dataset consisting of monthly data on global stock indices the BVAR model inherently incorporates co-movements in the stock markets. The time-varying specification of the covariance structure moreover accounts for sudden shifts in the level of volatility. In an out-of-sample forecasting application we show that the BVAR model with stochastic volatility significantly outperforms the random walk both in terms of root mean squared errors as well as Bayesian log predictive scores. The BVAR model without stochastic volatility, on the other hand, underperforms relative to the random walk. In a portfolio allocation exercise we moreover show that it is possible to use the forecasts obtained from our BVAR model with common stochastic volatility to set up simple investment strategies. Our results indicate that these simple investment schemes outperform a naive buy-and-hold strategy. (authors' abstract) / Series: Department of Economics Working Paper Series

A dissertação traz um breve resumo da história da siderurgia no Brasil, e comentários sobre o que vem ocorrendo no setor no mundo. A dissertação busca contribuir com a estimação de demanda para mercados oligopolizados, utilizando como proxie o mercado siderúrgico brasileiro. O objetivo, portanto, é estimar as variáveis econômicas que impactam o consumo de vergalhão no Brasil. Para isso são criados diversos modelos econométricos de demanda, utilizando as modelagens VAR (Vetor Auto-regressivo), BVAR (Vetor Autoregressivo Bayesiano) e Variáveis Instrumentais (IV). A metodologia BVAR foi aquela que apresentou os melhores resultados, com os seus coeficientes sendo robustos e estatisticamente significantes, além de reproduzirem a teoria econômica. / The present dissertation begins with a small resume of the history of the steel sector in Brazil, added to this some comments about that in the world. The dissertation aims to contribute with the estimation of oligopoly markets demand, using as a proxy the steel industry in Brazil. The main objective is to estimate the economics variables witch impact the rebar consumption in Brazil. The estimates are made using different econometric methodologies as VAR (Vector Autoregressive), BVAR (Bayesian Vector Autoregressive) and Instrumental Variables. The BVAR methodology is the one witch presents the best results, with the coefficients signs being robust and statistic significant, besides reproducing the economic theory.

A dissertação traz um breve resumo da história da siderurgia no Brasil, e comentários sobre o que vem ocorrendo no setor no mundo. A dissertação busca contribuir com a estimação de demanda para mercados oligopolizados, utilizando como proxie o mercado siderúrgico brasileiro. O objetivo, portanto, é estimar as variáveis econômicas que impactam o consumo de vergalhão no Brasil. Para isso são criados diversos modelos econométricos de demanda, utilizando as modelagens VAR (Vetor Auto-regressivo), BVAR (Vetor Autoregressivo Bayesiano) e Variáveis Instrumentais (IV). A metodologia BVAR foi aquela que apresentou os melhores resultados, com os seus coeficientes sendo robustos e estatisticamente significantes, além de reproduzirem a teoria econômica. / The present dissertation begins with a small resume of the history of the steel sector in Brazil, added to this some comments about that in the world. The dissertation aims to contribute with the estimation of oligopoly markets demand, using as a proxy the steel industry in Brazil. The main objective is to estimate the economics variables witch impact the rebar consumption in Brazil. The estimates are made using different econometric methodologies as VAR (Vector Autoregressive), BVAR (Bayesian Vector Autoregressive) and Instrumental Variables. The BVAR methodology is the one witch presents the best results, with the coefficients signs being robust and statistic significant, besides reproducing the economic theory.