O objetivo deste trabalho é comparar diferentes formas de estimação de modelos de apreçamento de ativos. Além dos métodos tradicionais, que utilizam toda a amostra no processo de estimação dos parâmetros do modelo, será utilizado o método rolling, que estima os parâmetros através da utilização de janelas móveis de tamanho fixo. Com isso, utilizando a técnica de backtesting, procura-se averiguar se o método rolling proporciona um ganho na qualidade de ajuste em modelos de apreçamento de ativos. / The aim of this project is to compare methods of estimating asset pricing models. In addition to using traditional methods, which estimate the models parameters by using the entire sample at once, the rolling method will be used. This method estimates the models parameters by using a rolling window of fixed size through the sample. By using backtesting, we seek to investigate whether the rolling approach provides an improvement in the goodness of fit in asset pricing models.

This paper develops and applies a Bayesian approach to Exploratory Factor Analysis that improves on ad hoc classical approaches. Our framework relies on dedicated factor models and simultaneously determines the number of factors, the allocation of each measurement to a unique factor, and the corresponding factor loadings. Classical identification criteria are applied and integrated into our Bayesian procedure to generate models that are stable and clearly interpretable. A Monte Carlo study confirms the validity of the approach. The method is used to produce interpretable low dimensional aggregates from a high dimensional set of psychological measurements. (authors' abstract)

This master thesis considers and evaluates a few different risk models for stock portfolios, including an ordinary sample covariance matrix, factor models and an approach inspired from random matrix theory. The risk models are evaluated by simulating minimum variance portfolios and employing a cross-validation. The Bloomberg+ transaction cost model is investigated and used to optimize portfolios of stocks, with respect to a trade off between the active risk of the portfolio and transaction costs. Further a few different simulations are performed while using the optimizer to rebalance long-only portfolios. The optimization problem is solved using an active-set algorithm. A couple of approaches are shown that may be used to visually try to decide a value for the risk aversion parameter λ in the objective function of the optimization problem. The thesis concludes that there is a practical difference between the different risk models that are evaluated. The ordinary sample covariance matrix is shown to not perform as well as the other models. It also shows that more frequent rebalancing is preferable to less frequent. Further the thesis goes on to show a peculiar behavior of the optimization problem, which is that the optimizer does not rebalance all the way to 0 in simulations, even if enough time is provided, unless it is explicitly required by the constraints.

This thesis wants to explore the forecasting potential of the multi-factor models to predict excess returns of the aggregated portfolio of the European stock mar- ket. These factors provided by Fama and French and Carhart are well-known in the field of asset pricing, we also add several financial and macroeconomic factors according to the literature. We establish a benchmark model of ARIMA and we compare the forecasting errors of OLS and the LSTM neural networks. Both models take the lagged excess returns and the inputs. We measure the performance with the root mean square error and mean absolute error. The results suggest that neural networks are in this particular task capable of bet- ter predictions given the same input as OLS but their forecasting error is not significantly lower according to the Diebold-Mariano test. JEL Classification C45, C53, C61, E37, G11, G15 Keywords Stocks, European market, Neural networks, LSTM, Factor Models, Fama-French, Predic- tions, RMSE Title Does LSTM neural network improve factor mod- els' predictions of the European stock market?

The dissertation consists of two essays. The first essay investigates the ability of prior returns, relative to some aggregate market returns, to predict future returns on industry style portfolios. By pooling time series of returns across industries for the period between July 1969 and June 2012, we find that prior returns differential predicts one month ahead returns negatively, even in the presence of a set of popular state variables. The predictability remains significant and negative for up to 5 month ahead returns. The predictability is shown to be robust to alternative specifications, estimation methodology and industry classifications. A possible explanation of this finding is based on time–varying (dynamic) loss aversion among investors. More specifically, when combined with house money effects, prior performance has inverse relationship with degree of loss aversion leading to predictability in the next period returns. The second essay examines the nature of time variation in the risk exposure of country mutual funds to the US market movement and to the benchmark foreign market movement. It uses weekly data on 15 closed end funds and 19 exchange traded funds for the sample period between January, 2001 and December, 2012. Conditional factor models are employed to uncover the time variation in the estimated betas through short horizon regressions. The findings of the paper indicate considerable time variation in risk exposure of country mutual funds to the US market and foreign market risk factors. Additional investigation reveals the following observations. First, the US market betas suffer greater variation over the sample period than the target foreign market betas. Second, the overall fluctuation in betas for the closed end funds is found to be higher than that for the exchange traded funds. Third, emerging market funds experience more oscillation in the risk exposure than their developed market counterparts. It is found that a combination of the US macroeconomic state variables and investors’ sentiment can predict future betas significantly. The findings of the paper have important implication for US investors seeking diversification benefits from country mutual funds.

Esse trabalho propõe estudar o grau de integração da estrutura a termo da taxa de juros com o mercado global e regional nos países da América Latina. Modelos de fatores dinâmicos foram usados para extrair os fatores globais, regionais e idiossincráticos da estrutura a termo como em Diebold, Li e Yue (2008) e Bae e Kim (2011). Foi encontrado que a estrutura a termo da taxa de juros da América Latina é integrada ao mercado global além de existir uma integração regional entre os países. Esse resultado é robusto ao fazer análises de subpériodos. No entanto, o proporção de variância explicada por cada fator varia conforme mudamos a amostra analisada. Essa variação pode ser consequência do período pós-crise e das politicas monetárias realizadas pelos principais Bancos Centrais no período. Ademais, a curva de juros do Brasil parece ter sido pouca influenciada por fatores globais pois o país apresentava condições macroeconômicas diferentes do restante do mundo. / In this work we propose to study the degree of integration of the term structure of interest rate of Latin America countries with global and regional markets. Using dynamic factor models as Diebold, Li e Yue (2008) and Bae e Kim (2011) to extract the global, regional and country specific factors we found that the term structure of interest rates of Latin America countries is integrated with global and regional markets. This result is robust studying different sample periods. However, the proportion of variance explained by those factors change when the sample periods change. This variation in the proportion of variance can be understood as consequence of the post crises period and the unconventional monetary policy that followed. Brazil term structure doesn\'t seem to be affected to global components. We interpret this last result as being a consequence of the different economic cycle that the country had comparing to the rest of the world.

No contexto dos modelos fatoriais existem diferentes metodologias para abordar a modelagem de séries temporais multivariadas que exibem uma estrutura não estacionária de segunda ordem, co- movimentos e transições no tempo. Modelos com mudanças estruturais abruptas e restrições rigorosas (muitas vezes irreais) nas cargas fatoriais, quando elas são funções determinísticas no tempo, foram propostos na literatura para lidar com séries multivariadas que possuem essas características. Neste trabalho, apresentamos um modelo fatorial com cargas variando continuamente no tempo para modelar séries temporais não estacionárias e um procedimento para sua estimação que consiste em dois estágios. No primeiro, os fatores latentes são estimados empregando os componentes principais das séries observadas. Em um segundo estágio, tratamos estes componentes principais como co-variáveis e as cargas funcionais são estimadas através de funções de ondaletas e mínimos quadrados generalizados. Propriedades assintóticas dos estimadores de componentes principais e de mínimos quadrados dos coeficientes de ondaletas são apresentados. O desempenho da metodologia é ilustrado através de estudos de simulação. Uma aplicação do modelo proposto no mercado spot de energia do Nord Pool é apresentado. / In the context of the factor models there are different methodologies to modeling multivariate time series that exhibit a second order non-stationary structure, co-movements and transitions over time. Models with abrupt structural changes and strict restrictions (often unrealistic) in factor loadings, when they are deterministic functions of time, have been proposed in the literature to deal with multivariate series that have these characteristics. In this work, we present a factor model with time-varying loadings continuously to modeling non-stationary time series and a procedure for its estimation that consists of two stages. First, latent factors are estimated using the principal components of the observed series. Second, we treat principal components obtained in first stage as covariate and the functional loadings are estimated by wavelet functions and generalized least squares. Asymptotic properties of the principal components estimators and least squares estimators of the wavelet coefficients are presented. The per- formance of the methodology is illustrated by simulations. An application to the model proposed in the energy spot market of the Nord Pool is presented.

No contexto dos modelos fatoriais existem diferentes metodologias para abordar a modelagem de séries temporais multivariadas que exibem uma estrutura não estacionária de segunda ordem, co- movimentos e transições no tempo. Modelos com mudanças estruturais abruptas e restrições rigorosas (muitas vezes irreais) nas cargas fatoriais, quando elas são funções determinísticas no tempo, foram propostos na literatura para lidar com séries multivariadas que possuem essas características. Neste trabalho, apresentamos um modelo fatorial com cargas variando continuamente no tempo para modelar séries temporais não estacionárias e um procedimento para sua estimação que consiste em dois estágios. No primeiro, os fatores latentes são estimados empregando os componentes principais das séries observadas. Em um segundo estágio, tratamos estes componentes principais como co-variáveis e as cargas funcionais são estimadas através de funções de ondaletas e mínimos quadrados generalizados. Propriedades assintóticas dos estimadores de componentes principais e de mínimos quadrados dos coeficientes de ondaletas são apresentados. O desempenho da metodologia é ilustrado através de estudos de simulação. Uma aplicação do modelo proposto no mercado spot de energia do Nord Pool é apresentado. / In the context of the factor models there are different methodologies to modeling multivariate time series that exhibit a second order non-stationary structure, co-movements and transitions over time. Models with abrupt structural changes and strict restrictions (often unrealistic) in factor loadings, when they are deterministic functions of time, have been proposed in the literature to deal with multivariate series that have these characteristics. In this work, we present a factor model with time-varying loadings continuously to modeling non-stationary time series and a procedure for its estimation that consists of two stages. First, latent factors are estimated using the principal components of the observed series. Second, we treat principal components obtained in first stage as covariate and the functional loadings are estimated by wavelet functions and generalized least squares. Asymptotic properties of the principal components estimators and least squares estimators of the wavelet coefficients are presented. The per- formance of the methodology is illustrated by simulations. An application to the model proposed in the energy spot market of the Nord Pool is presented.

This thesis is composed of three chapters which propose some novel approaches to model and forecast financial time series. The first chapter focuses on high frequency financial returns and proposes a quantile regression approach to model their intraday seasonality and dynamics. The second chapter deals with the problem of forecasting the yield curve including large datasets of macroeconomics information. While the last chapter addresses the issue of modelling the term structure of interest rates. The first chapter investigates the distribution of high frequency financial returns, with special emphasis on the intraday seasonality. Using quantile regression, I show the expansions and shrinks of the probability law through the day for three years of 15 minutes sampled stock returns. Returns are more dispersed and less concentrated around the median at the hours near the opening and closing. I provide intraday value at risk assessments and I show how it adapts to changes of dispersion over the day. The tests performed on the out-of-sample forecasts of the value at risk show that the model is able to provide good risk assessments and to outperform standard Gaussian and Student’s t GARCH models. The second chapter shows that macroeconomic indicators are helpful in forecasting the yield curve. I incorporate a large number of macroeconomic predictors within the Nelson and Siegel (1987) model for the yield curve, which can be cast in a common factor model representation. Rather than including macroeconomic variables as additional factors, I use them to extract the Nelson and Siegel factors. Estimation is performed by EM algorithm and Kalman filter using a data set composed by 17 yields and 118 macro variables. Results show that incorporating large macroeconomic information improves the accuracy of out-of-sample yield forecasts at medium and long horizons. The third chapter statistically tests whether the Nelson and Siegel (1987) yield curve model is arbitrage-free. Theoretically, the Nelson-Siegel model does not ensure the absence of arbitrage opportunities. Still, central banks and public wealth managers rely heavily on it. Using a non-parametric resampling technique and zero-coupon yield curve data from the US market, I find that the no-arbitrage parameters are not statistically different from those obtained from the Nelson and Siegel model, at a 95 percent confidence level. I therefore conclude that the Nelson and Siegel yield curve model is compatible with arbitrage-freeness.

This thesis consists of four chapters that investigate the performance and capital flows of hedge funds. The first two chapters of the thesis focus on hedge funds that have a pure emerging market strategy. Hedge funds should be well equipped to take advantage of opportunities in emerging markets due to their flexibility in investment strategy and lockup periods. However, the results show that, at the strategy level, emerging market hedge funds have only generated risk-adjusted returns in the most recent years of the sample period. Although emerging market hedge funds have performed poorly in the past, an important finding is the upward trend over time in performance. Given that other hedge fund strategies have a declining trend in alpha during the same period, the emerging market strategy may be where future alpha can be found. The third chapter investigates if there are capacity constraints in hedge fund strategies. The idea is that the alpha opportunities in the markets are limited. Thus, the more capital coming in to hedge funds, the higher competition for the investment opportunities. The findings reveal that mainly strategies that rely on liquidity in their underlying market show evidence of capacity constraints. That is, high past capital flows have a negative effect on current risk-adjusted returns. The last chapter investigates the out-of-sample performance of five allocation models relative to an equally weighted portfolio, when optimizing over hedge fund strategies. The findings show that for hedge fund investors the naive allocation model (1/N) with equal weights in each asset is not an efficient allocation. The risk-adjusted performance can be improved by using an optimal sample-based allocation model. Moreover, significant improvement in out-of-sample alpha can be made if the investor optimizes over non-systematic returns instead of total returns, which is an important results for investors seeking alpha. / <p>Diss. Stockholm : Handelshögskolan, 2008</p>