Made available in DSpace on 2010-04-20T21:00:11Z (GMT). No. of bitstreams: 4 Mauricio Mussashi Irie.pdf.jpg: 16022 bytes, checksum: d3dcf1b8020749a12b3baae53334cda5 (MD5) Mauricio Mussashi Irie.pdf.txt: 96735 bytes, checksum: 944036c3b3d1ae823db79daedb65dd6e (MD5) Mauricio Mussashi Irie.pdf: 7355110 bytes, checksum: 880246debd0d864c44768ebd4eaf2e6e (MD5) license.txt: 4886 bytes, checksum: 8fa2d810f5b64e058d76fb4986924cf0 (MD5) Previous issue date: 2009-02-06T00:00:00Z / Este trabalho explora com cuidado o lado específico da implementação de um modelo de alocação de ativos em que o risco é tratado de maneira integrada, não somente através do desvio padrão do portfólio, mas também considerando outras métricas de risco como, por exemplo, o Expected Shortfall. Além disso, utilizamos algumas técnicas de como trabalhar com as variáveis de modo a extrair do mercado os chamados "invariantes de mercado", fenômenos que se repetem e podem ser modelados como variáveis aleatórias independentes e identicamente distribuídas. Utilizamos as distribuições empíricas dos invariantes, juntamente com o método de Cópulas para gerar um conjunto de cenários multivariados simulados de preços. Esses cenários são independentes de distribuição, portanto são não paramétricos. Através dos mesmos, avaliamos a distribuição de retornos simulados de um portfólio através de um índice de satisfação que é baseado em uma função de utilidade quadrática e utiliza o Expected Shortfall como métrica de risco. O índice de satisfação incorpora o trade-off do investidor entre risco e retorno. Finalmente, escolhemos como alocação ótima aquela que maximiza o índice de satisfação ajustado a um parâmetro de aversão ao risco. Perseguindo esses passos, é possível obter um portfólio no qual a alocação em cada ativo, ou classe de ativos, reflete o prêmio esperado ao risco incorrido. / The present work carefully explores the implementation of an asset allocation model in which the risk measure considered is fully integrated, not only through the standard deviation for the portfolio, but also considering other risk metrics, for instance, the Expected Shortfall. Moreover, some statistical tools are used to extract from the market the so called “market invariants”, which are phenomena that tend to repeat themselves and can be modeled as i.i.d. random variables. We use the empirical distribution of the invariants, along with the Method of Copula to generate a set of simulated multivariate price scenarios. These scenarios are independent of distribution, therefore they are non-parametric. With these scenarios we evaluate the simulated return distribution of a portfolio through a satisfaction index which is based on a quadratic utility function and the risk measure considered is the Expected Shortfall. The satisfaction index summarizes the investor trade-off between risk and return. Finally, we choose the optimal allocation that maximizes the satisfaction index adjusted to a risk aversion parameter. In pursuing these steps, it is possible to obtain a portfolio in which the allocation of each asset class or security fully reflects the expected premium to the risk assumed.

Nesta tese discutimos abordagens que utilizam cópulas para descrever dependências entre instrumentos nanceiros e avaliamos a performance destes métodos. Muitas crises nanceiras aconteceram desde o nal da década de 90, incluindo a crise asiática (1997), a crise da dívida da Rússia (1998), a crise da bolha da internet (2000), as crises após o 9/11 (2001) e a guerra do Iraque (2003), a crise do subprime or crise nanceira global (2007-08), e a crise da dívida soberana europeia (2009). Todas estas crises levaram a uma perda maciça de riqueza nanceira e a um aumento da volatilidade observada, e enfatizaram a importância de uma política macroprudencial mais robusta. Em outras palavras, perturbações nanceiras tornam os processos econômicos altamente não-lineares, levando os principais bancos centrais a tomarem medidas contrárias para conter a angústia - nanceira. Devido aos complexos padrões de dependência dos mercados nanceiros, uma abordagem multivariada em grandes dimensões para a análise da dependência caudal é seguramente mais perspicaz do que assumir retornos com distribuição normal multivariada. Dada a sua exibilidade, as cópulas são capazes de modelar melhor as regularidades empiricamente veri cadas que são normalmente atribuídas a retornos nanceiros multivariados: (1) volatilidade condicional assimétrica com maior volatilidade para grandes retornos negativos e menor volatilidade para retornos positivos (HAFNER, 1998); (2) assimetria condicional (AIT-SAHALIA; BRANDT, 2001; CHEN; HONG; STEIN, 2001; PATTON, 2001); (3) excesso de curtose (TAUCHEN, 2001; ANDREOU; PITTIS; SPANOS, 2001); e (4) dependência temporal não linear (CONT, 2001; CAMPBELL; LO; MACKINLAY, 1997). A principal contribuição dos ensaios é avaliar se abordagens mais so sticadas do que o método da distância e o tradicional modelo de Markowitz podem tirar proveito de quaisquer anomalias/fricções de mercado. Os ensaios são uma tentativa de fornecer uma análise adequada destas questões usando conjuntos de dados abrangentes e de longo prazo. Empiricamente, demonstramos que as abordagens baseadas em cópulas são úteis em todos os ensaios, mostrando-se bené cas para modelar dependências em diferentes cenários, avaliando as medidas de risco caudais mais adequadamente e gerando rentabilidade superior a dos benchmarks utilizados. / In this thesis we discuss copula-based approaches to describe statistical dependencies within nancial instruments and evaluate its performance. Many nancial crises have occurred since the late 1990s, including the Asian crisis (1997), the Russian national debt crisis (1998), the dot-com bubble crisis (2000), the crises after 9-11 (2001) and Iraq war (2003), the subprime mortgage crisis or global nancial crisis (2007-08), and the European sovereign debt crisis (2009). All of these crises lead to a massive loss of nancial wealth and an upward in observed volatility and have emphasized the importance of a more robust macro-prudential policy. In other words, nancial disruptions make the economic processes highly nonlinear making the major central banks to take counter-measures in order to contain nancial distress. The methods for modeling uncertainty and evaluating the market risk on nancial markets are now under more scrutiny after the global nancial crisis. Due to the complex dependence patterns of nancial markets, a high-dimensional multivariate approach to tail dependence analysis is surely more insightful than assuming multivariate normal returns. Given its exibility, copulas are able to model better the empirically veri ed regularities normally attributed to multivariate nancial returns: (1) asymmetric conditional volatility with higher volatility for large negative returns and smaller volatility for positive returns (HAFNER, 1998); (2) conditional skewness (AITSAHALIA; BRANDT, 2001; CHEN; HONG; STEIN, 2001; PATTON, 2001); (3) excess kurtosis (TAUCHEN, 2001; ANDREOU; PITTIS; SPANOS, 2001); and (4) nonlinear temporal dependence (CONT, 2001; CAMPBELL; LO; MACKINLAY, 1997). The principal contribution of the essays is to assess if more sophisticated approaches than the distance method and plain Markowitz model can take advantage of any market anomalies/ fricctions. The essays are one attempt to provide a proper analysis in these issues using a long-term and comprehensive datasets. We empirically show that copula-based approaches are useful in all essays, proving bene cial to model dependencies in di erent scenarios, assessing the downside risk measures more adequately and yielding higher profitability than the benchmarks.

Nesta tese discutimos abordagens que utilizam cópulas para descrever dependências entre instrumentos nanceiros e avaliamos a performance destes métodos. Muitas crises nanceiras aconteceram desde o nal da década de 90, incluindo a crise asiática (1997), a crise da dívida da Rússia (1998), a crise da bolha da internet (2000), as crises após o 9/11 (2001) e a guerra do Iraque (2003), a crise do subprime or crise nanceira global (2007-08), e a crise da dívida soberana europeia (2009). Todas estas crises levaram a uma perda maciça de riqueza nanceira e a um aumento da volatilidade observada, e enfatizaram a importância de uma política macroprudencial mais robusta. Em outras palavras, perturbações nanceiras tornam os processos econômicos altamente não-lineares, levando os principais bancos centrais a tomarem medidas contrárias para conter a angústia - nanceira. Devido aos complexos padrões de dependência dos mercados nanceiros, uma abordagem multivariada em grandes dimensões para a análise da dependência caudal é seguramente mais perspicaz do que assumir retornos com distribuição normal multivariada. Dada a sua exibilidade, as cópulas são capazes de modelar melhor as regularidades empiricamente veri cadas que são normalmente atribuídas a retornos nanceiros multivariados: (1) volatilidade condicional assimétrica com maior volatilidade para grandes retornos negativos e menor volatilidade para retornos positivos (HAFNER, 1998); (2) assimetria condicional (AIT-SAHALIA; BRANDT, 2001; CHEN; HONG; STEIN, 2001; PATTON, 2001); (3) excesso de curtose (TAUCHEN, 2001; ANDREOU; PITTIS; SPANOS, 2001); e (4) dependência temporal não linear (CONT, 2001; CAMPBELL; LO; MACKINLAY, 1997). A principal contribuição dos ensaios é avaliar se abordagens mais so sticadas do que o método da distância e o tradicional modelo de Markowitz podem tirar proveito de quaisquer anomalias/fricções de mercado. Os ensaios são uma tentativa de fornecer uma análise adequada destas questões usando conjuntos de dados abrangentes e de longo prazo. Empiricamente, demonstramos que as abordagens baseadas em cópulas são úteis em todos os ensaios, mostrando-se bené cas para modelar dependências em diferentes cenários, avaliando as medidas de risco caudais mais adequadamente e gerando rentabilidade superior a dos benchmarks utilizados. / In this thesis we discuss copula-based approaches to describe statistical dependencies within nancial instruments and evaluate its performance. Many nancial crises have occurred since the late 1990s, including the Asian crisis (1997), the Russian national debt crisis (1998), the dot-com bubble crisis (2000), the crises after 9-11 (2001) and Iraq war (2003), the subprime mortgage crisis or global nancial crisis (2007-08), and the European sovereign debt crisis (2009). All of these crises lead to a massive loss of nancial wealth and an upward in observed volatility and have emphasized the importance of a more robust macro-prudential policy. In other words, nancial disruptions make the economic processes highly nonlinear making the major central banks to take counter-measures in order to contain nancial distress. The methods for modeling uncertainty and evaluating the market risk on nancial markets are now under more scrutiny after the global nancial crisis. Due to the complex dependence patterns of nancial markets, a high-dimensional multivariate approach to tail dependence analysis is surely more insightful than assuming multivariate normal returns. Given its exibility, copulas are able to model better the empirically veri ed regularities normally attributed to multivariate nancial returns: (1) asymmetric conditional volatility with higher volatility for large negative returns and smaller volatility for positive returns (HAFNER, 1998); (2) conditional skewness (AITSAHALIA; BRANDT, 2001; CHEN; HONG; STEIN, 2001; PATTON, 2001); (3) excess kurtosis (TAUCHEN, 2001; ANDREOU; PITTIS; SPANOS, 2001); and (4) nonlinear temporal dependence (CONT, 2001; CAMPBELL; LO; MACKINLAY, 1997). The principal contribution of the essays is to assess if more sophisticated approaches than the distance method and plain Markowitz model can take advantage of any market anomalies/ fricctions. The essays are one attempt to provide a proper analysis in these issues using a long-term and comprehensive datasets. We empirically show that copula-based approaches are useful in all essays, proving bene cial to model dependencies in di erent scenarios, assessing the downside risk measures more adequately and yielding higher profitability than the benchmarks.

Submitted by Hugo Rego (hl_freitas@yahoo.com.br) on 2012-05-20T18:47:32Z No. of bitstreams: 1 Dissertação de Mestrado_Hugo L F de Moraes Rêgo.pdf: 699903 bytes, checksum: 7d039f1507408214660b09b7998f05b3 (MD5) / Approved for entry into archive by Gisele Isaura Hannickel (gisele.hannickel@fgv.br) on 2012-05-21T12:11:12Z (GMT) No. of bitstreams: 1 Dissertação de Mestrado_Hugo L F de Moraes Rêgo.pdf: 699903 bytes, checksum: 7d039f1507408214660b09b7998f05b3 (MD5) / Made available in DSpace on 2012-05-21T12:30:46Z (GMT). No. of bitstreams: 1 Dissertação de Mestrado_Hugo L F de Moraes Rêgo.pdf: 699903 bytes, checksum: 7d039f1507408214660b09b7998f05b3 (MD5) Previous issue date: 2012-04-19 / The goal of this work was to compare two different asset allocation methodologies, the classic and the Bayesian one. The utilized model was that of Meucci (2005). In order to reach this goal, empirical exercises were performed, utilizing data from the Brazilian financial market. The results found indicate that the Bayesian asset portfolio outperformed the classic one in terms of return and volatility, whereas the classic portfolio outperformed the market index. Moreover, this work also comprises modifications in the prior utilized in the Bayesian estimation. / Este trabalho teve como objetivo comparar duas metodologias de alocação ótima de ativos, a metodologia clássica e a metodologia bayesiana. O modelo utilizado foi o de Meucci (2005). Foram realizados diversos exercícios empíricos de montagem de carteiras de ativos seguindo essas metodologias, utilizando para isso dados do mercado acionário brasileiro. Os resultados encontrados indicam uma superioridade de desempenho, tanto em termos de retorno quanto de volatilidade, da carteira bayesiana em relação à clássica e desta em relação ao índice de mercado. Ademais, o trabalho também compreende modificações na prior utilizada na estimação bayesiana.

Submitted by Pedro da Costa Neves Neto (pedrocnn@gmail.com) on 2012-09-04T03:40:26Z No. of bitstreams: 1 Dissertacao_PCNN_final.pdf: 2144117 bytes, checksum: 6985af49978cc36281a6038742aa9c6f (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br) on 2012-09-04T12:47:22Z (GMT) No. of bitstreams: 1 Dissertacao_PCNN_final.pdf: 2144117 bytes, checksum: 6985af49978cc36281a6038742aa9c6f (MD5) / Made available in DSpace on 2012-09-04T12:49:51Z (GMT). No. of bitstreams: 1 Dissertacao_PCNN_final.pdf: 2144117 bytes, checksum: 6985af49978cc36281a6038742aa9c6f (MD5) Previous issue date: 2012-08-24 / This study aimed at evaluating the intertemporal hedging demand for stocks and bonds to investors in Brazil and the U.S. Following the framework of the dynamic portfolio choice theory established by Merton (1969, 1971, 1973) and Samuelson (1969), and the multivariate model proposed by Campbell, Chan and Viceira (2003), we find the optimal portfolio choice for infinitely-lived investors with Epstein-Zin-Weil utility defined over consumption who allocate their wealth in stocks, long-term bonds and short-term bills, whose returns are described by a first order vector autoregressive process. There is evidence of positive intertemporal hedging demand for stocks and bonds to U.S. investors, but the hedging demand for bonds is greater than for stocks, what contrasts with the results found in Campbell, Chan and Viceira (2003). The selected period, from January 1998 to March 2012, witnessed lower stock returns, lower dividend yield predictive power and a strong downward trend in short-term interest rates, which may have influenced the results. Risk averse brazilian investors also protect themselves against shocks in future investment opportunities in long-term bonds, meaning that this acts like the riskless asset for brazilian long-term investors during periods of low inflation risk. The intertemporal hedging demand for stocks is low, but positive. However, some facts indicate that stocks may play a stronger role in hedging demand in the near future, following the role they already play in economies with developed financial markets. Assets with hedging demand characteristics exhibit the desirable property of decreasing variance across time, which is very important to long-term investment policy. These are important results for pension system players and all those who are interested in long-term asset allocation. / Este estudo teve por objetivo mensurar a demanda por proteção intertemporal por ações e títulos longos de renda fixa, no Brasil e nos EUA. Seguindo o arcabouço da teoria de escolha dinâmica de carteiras de Merton (1969, 1971, 1973) e Samuelson (1969), e o modelo multivariado proposto por Campbell, Chan e Viceira (2003), encontramos a solução ótima de carteira para investidores de longo prazo com função utilidade Epstein-Zin-Weil sobre uma corrente de consumo sem data terminal, e que alocam entre ações e títulos de curto e longo prazo, cujos retornos são representados por um vetor autorregressivo de primeira ordem. Encontramos evidência de demanda positiva por proteção intertemporal por ações e títulos de longo prazo para o investidor americano, porém a demanda por títulos é muito superior à por ações, resultado que difere de Campbell, Chan e Viceira (2003). A escolha do período utilizado, de janeiro de 1998 a março de 2012, marcado por menor retorno das ações, menor poder preditivo do dividend yield e clara trajetória de queda na taxa de juros de curto prazo, teve influência no resultado. Investidores brasileiros avessos a risco se protegem da deterioração nas oportunidades de investimento em títulos de longo prazo, mostrando que estes agem como ativos livres de risco no Brasil em períodos de inflação controlada. A demanda por proteção intertemporal por ações é de baixa magnitude, mas positiva, e estão presentes indícios de que possam convergir, ao longo do tempo, para o papel que exercem para investidores nas economias com mercados financeiros desenvolvidos. Pela característica de proteção intertemporal, ações e títulos de longo prazo contém a desejável propriedade de variância acumulada decrescente, em períodos longos de investimento. É um resultado a ser considerado por participantes de sistemas previdenciários e todos os interessados na alocação de ativos de longo prazo.

[pt] O problema de como alocar ativos de forma eficiente tem sido uma das questões fundamentais em Finanças. Uma das mudanças recentes no mercado brasileiro tem sido o crescimento dos fundos de Contribuição Definida, seguindo a tendência observada em outros mercados. Entretanto, ao focalizar decisões de investimento sob o ponto de vista do investidor individual, surge a necessidade de incluir a tributação no processo de alocação de carteiras. Neste contexto, este trabalho analisa a decisão de alocação e localização preferencial para as classes de ativos, em veículos de investimento com tributação convencional, ou com diferimento de imposto, mediante a legislação local. O investidor possui dois tipos de riqueza, seu capital financeiro acumulado, e o capital humano, representado pela capacidade de gerar rendimentos futuros. A solução é obtida por alocação multiperiódica de recursos, seguindo o critério de maximização da utilidade esperada da riqueza final. Face à eficiência tributária dos fundos mútuos de ações domésticos, estes podem ser priorizados na localização externa aos planos com tributação diferida, coerente com resultados recentes para o mercado americano. Contudo, se existem diferenças nas rentabilidades das classes de ativos, nos distintos veículos de investimentos, aquela localização prioritária pode mudar, pelo menos para aplicações com objetivos a serem atingidos em prazos reduzidos. / [en] The question of how to allocate assets efficiently has been one of the central issues in Finance. As perceived in other markets, one of the recent trends in the Brazilian market has been the growth of Defined Contribution Funds. However, when focusing on investment decisions for individual investors, taxes must be taken into account. In this context, the asset allocation and location is solved for brazilian assets, when the investor has to save in both investment vehicles with conventional, and deferred taxation, according to the local rules. The investor has two kinds of wealth, the accumulated financial wealth, and the human capital, representing the cash-flows that can be produced in the future. The solution is obtained through multi-period asset allocation, for an investor maximizing the expected utility of terminal wealth. Due to the tax efficiency of domestic equity mutual funds, stocks should have preferential location outside the deferred account, in accordance with recent results for the american market. However, if there are performance differences among the asset classes, within distinct investment vehicles, that preferred location may change, at least for short term investment objectives.

La présente thèse s’intéresse aux modèles d’allocation stratégiques d’actifs et à leurs applications pour la gestion des réserves financières des régimes de retraite par répartition, en particulier ceux partiellement provisionnés. L’étude de l’utilité des réserves pour un système par répartition et a fortiori de leur gestion reste un sujet peu exploré. Les hypothèses classiques sont parfois jugées trop restrictives pour décrire l'évolution complexe des réserves. De nouveaux modèles et de nouveaux résultats sont développés à trois niveaux : la génération de scénarios économiques (GSE), les techniques d’optimisation numérique et le choix de l’allocation stratégique optimale dans un contexte de gestion actif-passif (ALM). Dans le cadre de la génération de scénarios économiques et financiers, certains indicateurs de mesure de performance du GSE ont été étudiés. Par ailleurs, des améliorations par rapport à ce qui se pratique usuellement lors de la construction du GSE ont été apportées, notamment au niveau du choix de la matrice de corrélation entre les variables modélisées. Concernant le calibrage du GSE, un ensemble d’outils permettant l’estimation de ses différents paramètres a été présenté. Cette thèse a également accordé une attention particulière aux techniques numériques de recherche de l'optimum, qui demeurent des questions essentielles pour la mise en place d'un modèle d'allocation. Une réflexion sur un algorithme d’optimisation globale d’une fonction non convexe et bruitée a été développée. L’algorithme permet de moduler facilement, au moyen de deux paramètres, la réitération de tirages dans un voisinage des points solutions découverts, ou à l’inverse l’exploration de la fonction dans des zones encore peu explorées. Nous présentons ensuite des techniques novatrices d'ALM basées sur la programmation stochastique. Leur application a été développée pour le choix de l’allocation stratégique d’actifs des régimes de retraite par répartition partiellement provisionnés. Une nouvelle méthodologie pour la génération de l’arbre des scénarios a été adoptée à ce niveau. Enfin, une étude comparative du modèle d’ALM développé avec celui basé sur la stratégie Fixed-Mix a été effectuée. Différents tests de sensibilité ont été par ailleurs mis en place pour mesurer l’impact du changement de certaines variables clés d’entrée sur les résultats produits par notre modèle d’ALM / This thesis focuses on the strategic asset allocation models and on their application for the financial reserve management of a pay-as-you-go (PAYG) retirement schemes, especially those with partial provision. The study of the reserve utility for a PAYG system and of their management still leaves a lot to be explored. Classical hypothesis are usually considered too restrictive for the description of the complex reserve evolution. New models and new results have been developed over three levels : economic scenario generation (ESG), numerical optimization techniques and the choice of optimal strategic asset allocation in the case of an Asset-Liability Management (ALM). For the generation of financial and economic scenarios, some ESG performance indicators have been studied. Also, we detailed and proposed to improve ESG construction, notably the choice of the correlation matrix between modelled variables. Then, a set of tools were presented so that we could estimate ESG parameters variety. This thesis has also paid particular attention to numerical techniques of optimum research, which is an important step for the asset allocation implementation. We developed a reflexion about a global optimisation algorithm of a non convex and a noisy function. The algorithm allows for simple modulating, through two parameters, the reiteration of evaluations at an observed point or the exploration of the noisy function at a new unobserved point. Then, we presented new ALM techniques based on stochastic programming. An application to the strategic asset allocation of a retirement scheme with partial provision is developed. A specific methodology for the scenario tree generation was proposed at this level. Finally, a comparative study between proposed ALM model and Fixed-Mix strategy based model was achieved. We also made a variety of a sensitivity tests to detect the impact of the input values changes on the output results, provided by our ALM model

This thesis investigates the impact of Solvency II on the asset side of Swedish mutual life insurers. With the help of a quantitative analysis and a qualitative examination of our results we find that there will be a significant change in demand for certain products. A substantial increase in demand for government bonds and interest rate swaps with long maturities should be expected. Furthermore, both corporate and covered bonds will be more attractive investments under the new regulatory framework. Another big impact is the lower risk-adjusted return for equity, which over time will lead to a reduction in Swedish life insurers’ relatively high exposure to equity and equity based products. Furthermore, we conclude that there are large gains to be made by incorporating an optimization with regard to the solvency capital requirements dictated by the legislative texts. / Denna uppsats har undersökt vilken inverkan Solvens II kommer ha på svenska ömsesidiga livbolags tillgångssidor. Med hjälp av en kvantitativ analys och en kvalitativ undersökning av våra resultat har vi funnit att det kommer ske en betydande förändring i efterfrågan av vissa instrument. En stor ökning på efterfrågan av statsobligationer och ränteswappar med långa löptider är att vänta. Dessutom kommer både företags- och säkerställda obligationer vara betydligt mer attraktiva investeringsalternativ under det nya regelverket. En annan stor inverkan är den lägre riskjusterade avkastningen för aktier och aktierelaterade produkter. Över tid kommer detta sannolikt leda till en reduktion av svenska livbolags relativt höga exponering har gentemot aktier. Därutöver finner vi att bolagen har mycket att vinna på att införliva en optimering med avseende på de av regelverket angivna kapitalkraven.

This thesis investigates the two investment strategies Dogs of Dow and The Magic Formula. We test how the strategies perform when getting weighted to ESG scores and also if they outperform OMXSPI during the years 2012-2022. What we find in our study is that when returns are risk adjusted, both Dogs of Dow and The Magic Formula and their ESG weighted portfolios outperform the benchmark during the period. We also conclude that ESG weighted portfolios yield lower returns than equally weighted Dogs of Dow and The Magic Formula portfolios. The portfolio that produce the highest return was the equally weighted Dogs of Dow portfolio. For the value at risk we find that on a five percent significant level, the portfolios observe values from -1,55% to -1,69%.

Le premier chapitre montre que les rendements des fonds investis dans des actions illiquides (“fonds illiquides”) sont mieux que ceux des fonds investis dans des actions liquides. Cette différence provient des capacités de fonds illiquides de sélectionner les actions. Les actions détenues par les fonds illiquides ont une meilleure performance que des portefeuilles qui ont les mêmes caractéristiques. Les fonds liquides déclarent des indices de référence par rapport auxquels leurs rendements sont plus importants. Un portefeuille d’actions détenues par les fonds illiquides a une meilleure performance qu’un portefeuille d’actions détenues par les fonds liquides. Le second chapitre documente une prédictibilité des rendements. Dans ce chapitre, les périodes d’opportunités sont les périodes où les rendements des actions faisant l’objet d’une analyse régulière par les analystes (les actions suivies) s'écartent de ceux des actions qui ne sont pas suivie (les actions négligés). Les rendements ultérieurs des actions faciles à évaluer sont plus importants quand les opportunités étaient grandes, par rapport aux périodes où les opportunités étaient limitées. Ce comportement est cohérent avec un modelé où les investisseurs exigent une prime pour supporter le risque de sélection défavorable. Le troisième chapitre explore les moments où les fonds d’investissement changent leur style d’investissement (le style est défini comme exposition au risque, prenant en compte les facteurs de risque habituels). Les fonds ne prennent pas plus des risques quand il serait plus rentable de le faire. Après avoir eu des mauvais rendements, les fonds se rapprochent du style des fonds similaires, mais qui ont eu des bons rendements. Le style de jeunes fonds s’écarte du style de fonds anciens. Les nouveaux gérants des fonds s’écartent du style de fonds avec des anciens gérants. Quand un fond prend plus de risque d’une côté, il n’essaye pas d’aborder systématiquement les autres côtés du risque. / The first chapter shows that mutual funds that hold illiquid stocks (“illiquid funds”) outperform funds that hold liquid stocks (“liquid funds”). There is evidence this outperformance arises from stock selection skills of illiquid funds. The stocks held by illiquid funds outperform portfolios matched by characteristics. Liquid funds declare benchmarks that make their benchmarkadjusted returns appear larger. A portfolio of stocks held by illiquid funds subsequently outperforms a portfolio of stocks held by liquid funds. The second chapter documents a predictability pattern in returns. This chapter identifies high opportunities in stocks with difficult valuation as times when returns of neglected stocks diverge from returns of covered stocks. Subsequent returns of stocks with difficult valuation are higher when beginning of period opportunities are high, as compared to when beginning of period opportunities are low. This is consistent with an information risk theory, where investors demand a higher premium to hold stocks with higher probability of informed trading, because they fear adverse selection. The third chapter explores instances when mutual funds change their style (style is regarded as risk exposure alongside usual factors). Mutual funds do not take more risk when it is more profitable to do so. After performing badly, mutual funds move closer to the style of good performing peer funds. Young funds' styles diverge from the style of old peer funds. Recently hired managers diverge in style from veteran managers of peer funds. When the average fund takes more risk alongside a style dimension, it does not simultaneously consider other style dimensions.