We evaluate the validity of the Adaptive Market Hypothesis (AMH) in a Swedish context by testing for stock return predictability on the OMXS30 stock index between 1986 and 2014 using daily returns and monthly two year moving subsamples. To our knowledge, this is the first study to evaluate the AMH in a Swedish context. Three tests for linear independence based on Lo and MacKinlay (1988) variance ratio test, namely the Chow and Denning joint test as well as Wright (2000) joint rank and sign tests are used. We also test for non-linear independence using the BDS test statistics. Presented in our findings is evidence of time-varying predictability where stock returns go through periods of return predictability and non-predictability. When evaluating the different market conditions (volatility, bull, bear, up, down and normal markets) we find that these different market conditions govern the degree of stock return predictability in different ways. Our findings support the AMH on the OMXS30 stock index and in contrast to previous research regarding market efficiency on the Swedish stock market, we do not find persistent stock return predictability over the short and long term.

This thesis consists of four self-contained papers related to the change of market structure and the quality of equity market. In Paper [I] we found, by using of a Flexible Dynamic Component Correlations (FDCC) model, that the creation of a common cross-border stock trading platform has increased the long-run trends in conditional correlations between foreign and domestic stock market returns. In Paper [II] we study whether the creation of a uniform Nordic and Baltic stock trading platform has affected weak-form information efficiency. The results indicate that the stock market consolidations have had a positive effect on the information efficiency and turnover for an average firm. The merger effects are, however, asymmetrically distributed in the sense that relatively large (small) firms located on relatively large (small) markets experience an improved (reduced) information efficiency and turnover. Although the results indicate that changes in the level of investor attention (measured by turnover) may explain part of the changes in information efficiency, they also lend support to the hypothesis that merger effects may partially be driven by changes in the composition of informed versus uninformed investors following a stock. Paper [III] analyzes whether the measured level of trust in different countries can explain bilateral stock market correlations. One finding is that generalized trust among nations is a robust predictor for stock market correlations. Another is that the trust effect is larger for countries which are close to each other. This indicates that distance mitigates the trust effect. Finally, we confirm the effect of trust upon stock market correlations, by using particular trust data (bilateral trust between country A and country B) as an alternative measurement of trust. In Paper [IV] we present the impact of the stock market mergers that took place in the Nordic countries during 2000 – 2007 on the probabilities for stock price jumps, i.e. for relatively extreme price movements. The main finding is that stock market mergers, on average, reduce the likelihood of observing stock price jumps. The effects are asymmetric in the sense that the probability of sudden price jumps is reduced for large and medium size firms whereas the effect is ambiguous for small size firms. The results also indicate that the market risk has been reduced after the stock market consolidations took place.

The predictability of the US-based international mutual fund returns has received renewed consideration in recent academic studies. This dissertation extends recent research by exploring the 2,479 daily return observations covering the period from January 4, 1993 to October 31, 2002 for all categories of international mutual funds. This exploration splits the sample, uses the initial sub-sample to investigate return patterns of international mutual funds and develops trading rules based on the predictable return patterns, and tests those rules on the holdout sample. The empirical findings suggest that smart investors may earn higher riskadjusted returns by following daily dynamic trading strategies. The excess returns earned by investors are statistically and economically significant, irrespective of load or no-load mutual funds and even in the presence of various exchange restrictions and regulations.

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.

This thesis consists of three essays. In the first essay, I show that information about dividends implied in derivative markets predicts future dividend growth and thereby improves the forecasts of short-run returns on the aggregate market. In the second essay, we analyze the impact of options trading on the price distribution of the underlying asset. Specifically, we show that S&P 500 futures finish in the proximity of the closest strike price on days when options on S&P 500 futures expire. We document that this effect is mainly driven by the rebalancing of delta hedges of the market maker. In the third essay, we develop a theory of price support in security markets that arises from conflict of interests, and we test our hypothesis in the context of the Spanish mutual fund industry. In particular, we analyze how bank-affiliated mutual funds trade in the stock of the parent bank and show that, consistently with the price support hypothesis, affiliated mutual funds tend to increase their holdings of the parent bank’s stock following a large drop in its price. / Esta tesis consta de tres capítulos. En el primer capítulo, muestro que la información sobre dividendos implícita en los mercados de derivados predice el crecimiento futuro de los dividendos, mejorando así las predicciones de los rendimientos a corto plazo en el mercado agregado. En el segundo capítulo, analizamos el impacto de la compraventa de opciones en la distribución del precio del activo subyacente. En concreto, mostramos que los futuros del S&P 500 terminan en el entorno del precio de ejercicio más próximo en los días en que las opciones sobre los futuros del S&P 500 expiran. Documentamos que este efecto está principalmente motivado por el reajuste de la cobertura delta de los intermediarios. En el tercer capítulo, desarrollamos una teoría de sostenimiento de precios en los mercados de valores motivado por un conflicto de intereses y testamos nuestra hipótesis en el contexto de la industria española de fondos de inversión. En concreto, analizamos cómo los fondos de inversión afiliados a un banco operan las acciones del banco matriz y mostramos que, consecuentemente con la hipótesis del sostenimiento de precios, los fondos de inversión filiales tienden a incrementar sus posiciones en las acciones del banco matriz después de una caída importante de su cotización.

[pt] Essa tese é composta por três artigos em econometria financeira. Os dois primeiros artigos exploram a relação entre retornos intradiários do mercado de equities e a implied volatility, representada pelo Índice de Volatilidade da CBOE (VIX). Nos dois artigos, estimamos previsões um minuto à frente utilizando janelas rolantes para cada dia. No primeiro artigo, as estimativas indicam que nossos modelos de fatores de volatilidade têm uma performance superior a benchmarks tradicionais em uma análise de séries de tempo em alta frequência, mesmo aos excluirmos períodos de crise da amostra. Os resultados também indicam uma performance fora da amostra maior para dias em que não ocorrem anúncios macroeconômicos. A performance é ainda maior quando removemos períodos de crise. O segundo artigo propõe uma abordagem de aprendizado de máquinas para modelar esse exercício de previsão. Implementamos um método de estimação intradiário minuto a minuto com janelas móveis, utilizando dois tipos de modelos não lineares: redes neurais com Long-Short-Term Memory (LSTM) e Random Forests (RF). Nossas estimativas mostram que o VIX é o melhor previsor de retornos de mercado intradiários entre os candidatos na nossa análise, especialmente quando implementadas através do modelo LSTM. Esse modelo também melhora significativamente a performance quando utilizamos o retorno de mercado defasado como variável preditiva. Finalmente, o último artigo explora uma extensão multivariada do método FarmPredict, combinando modelos vetoriais autoregressivos aumentados em fatores (FAVAR) e modelos esparsos em um ambiente de alta dimensão. Utilizando um procedimento de três estágios, somos capazes de estimar e prever fatores e seus loadings, que podem ser observados, não observados ou ambos, assim como uma estrutura idiossincrática fracamente esparsa. Realizamos uma aplicação dessa metodologia em um painel de volatilidades realizadas e os resultados de performance do método em etapas indicam melhorias quando comparado a benchmarks consolidados. / [en] This thesis is composed of three papers on financial econometrics. The first two papers explore the relation between intraday equity market returns and implied volatility, represented by the CBOE Volatility Index (VIX). In both papers, we estimate one-minute-ahead forecasts using rolling windows within a day. In the first paper, the estimates indicate that our volatility factor models outperform traditional benchmarks at high frequency time-series analysis, even when excluding crisis periods. We also find that the model has a better out-of-sample performance at days without macroeconomic announcements. Interestingly, these results are amplified when we remove the crisis period. The second paper proposes a machine learning modeling approach to this forecasting exercise. We implement a minute-by-minute rolling window intraday estimation method using two nonlinear models: Long-Short-Term Memory (LSTM) neural networks and Random Forests (RF). Our estimations show that the VIX is the strongest candidate predictor for intraday market returns in our analysis, especially when implemented through the LSTM model. This model also improves significantly the performance of the lagged market return as predictive variable. Finally, the third paper explores a multivariate extension of the FarmPredict method, by combining factor-augmented vector autoregressive (FAVAR) and sparse models in a high-dimensional environment. Using a three-stage procedure, we estimate and forecast factors and its loadings, which can be observed, unobserved, or both, as well as a weakly sparse idiosyncratic structure. We provide an application of this methodology to a panel of daily realized volatilities. Finally, the accuracy of the stepwise method indicates improvements of this forecasting method when compared to consolidated benchmarks.

Multi-period portfolio optimization (MPO) has gained a lot of interest in modern portfolio theory due to its consideration for inter-temporal trading e effects, especially market impacts and transactions costs, and for its subtle reliability on return predictability. However, because of the heavy computational demand, portfolio policies based on this approach have been sparsely explored. In that regard, a tractable MPO framework proposed by N. Gârleanu & L. H. Pedersen has been investigated. Using the stochastic control framework, the authors provided a closed form expression of the optimal policy. Moreover, they used a specific, yet flexible return predictability model. Excess returns were expressed using a linear factor model, and the predicting factors were modeled as mean reverting processes. Finally, transactions costs and market impacts were incorporated in the problem formulation as a quadratic function. The elaborated methodology considered that the market returns dynamics are governed by fast and slow mean reverting factors, and that the market transactions costs are not necessarily quadratic. By controlling the exposure to the market returns predicting factors, the aim was to uncover the importance of the mean reversion speeds in the performance of the constructed trading strategies, under realistic market costs. Additionally, for the sake of comparison, trading strategies based on a single-period mean variance optimization were considered. The findings suggest an overall superiority in performance for the studied MPO approach even when the market costs are not quadratic. This was accompanied with evidence of better usability of the factors' mean reversion speed, especially fast reverting factors, and robustness in adapting to transactions costs. / Portföljoptimering över era perioder (MPO) har fått stort intresse inom modern portföljteori. Skälet till detta är att MPO tar hänsyn till inter-temporala handelseffekter, särskilt marknadseffekter och transaktionskostnader, plus dess tillförlitlighet på avkastningsförutsägbarhet. På grund av det stora beräkningsbehovet har dock portföljpolitiken baserad på denna metod inte undersökts mycket. I det avseendet, har en underskriven MPO ramverk som föreslagits av N.Gârleanu L. H. Pedersen undersökts. Med hjälp av stokastiska kontrollramen tillhandahöll författarna formuläret för sluten form av den optimala politiken. Dessutom använde de en specifik, men ändå flexibel returförutsägbarhetsmodell. Överskjutande avkastning uttrycktes med hjälp av en linjärfaktormodell och de förutsägande faktorerna modellerades som genomsnittligaåterföringsprocesser. Slutligen inkorporerades transaktionskostnader och marknadseffekter i problemformuleringen som en kvadratisk funktion. Den utarbetade metodiken ansåg att marknadens avkastningsdynamik styrs av snabba och långsammaåterhämtningsfaktorer, och att kostnaderna för marknadstransaktioner inte nödvändigtvis är kvadratiska. Genom att reglera exponeringen mot marknaden återspeglar förutsägande faktorer, var målet att avslöja vikten av de genomsnittliga omkastningshastigheterna i utförandet av de konstruerade handelsstrategierna, under realistiska marknadskostnader. Dessutom, för jämförelses skull, övervägdes handelsstrategier baserade på en enstaka genomsnittlig variansoptimering. Resultaten tyder på en överlägsen överlägsenhet i prestanda för det studerade MPO-tillvägagångssättet, även när marknadsutgifterna inte är kvadratiska. Detta åtföljdes av bevis för bättre användbarhet av faktorernas genomsnittliga återgångshastighet, särskilt snabba återställningsfaktorer och robusthet vid anpassning till transaktionskostnader

[pt] Esta tese é composta por dois ensaios sobre finanças empíricas. O primeiro se concentra nos mercados de câmbio e apresenta medidas de mudanças na inclinação da estrutura de curto prazo das taxas de juros para os EUA e outros países de G10, usando contratos de futuros de 3 e 6 meses. Essas mudanças na inclinação têm impacto imediato nos retornos da moeda e também forte efeito retardado nas semanas seguintes, o que implica que as moedas são previsíveis tanto dentro quanto fora da amostra. Os investidores que condicionam na inclinação para negociar taticamente uma carteira comprada em moedas G10 contra o Dólar americano melhoram os índices de Sharpe para 0,4-0,9, em relação a 0,15 de uma estratégia de buy and hold. Uma carteira de moeda neutra em dólares que classifica as moedas dos países do G10 de acordo com a inclinação no cross-section também oferece índices de Sharpe mais altos do que outras estratégias de moeda como o carry trade. Essas descobertas são compatíveis com uma reação defasada do mercado de câmbio às informações sobre taxas de juros. O segundo ensaio propõe uma nova medida que usa apenas informações de dispersão cross-section de betas do modelo CAPM para prever retornos agregados de mercado para os EUA. Esta escolha de preditores é baseada em argumentos teóricos simples de que as medidas associadas à dispersão dos betas do CAPM, em alguns cenários, devem ser relacionadas aos retornos futuros de mercado esperados. Essas medidas de dispersão de fato prevêem o prêmio de risco de mercado em vários horizontes e fornecem alto poder preditivo dentro e fora da amostra. O R2 fora da amostra atinge até 10 porcento na frequência anual (0,7 porcento mensal) e são robustos a diferentes janelas de estimação. Ao contrário da maioria das medidas encontradas na literatura, a nossa não é baseado em preço ou valuation ratios. Nossas medidas variam com o ciclo econômico e se correlacionam com outras variáveis de previsão comumente usadas, como razões de dividendo-preço e consumo-riqueza, mas fornecem poder explicativo acima e além dos preditores padrão. Nossos resultados fornecem evidências adicionais de que a dispersão dos betas ao longo do tempo é função da variação temporal do prêmio de risco de mercado. / [en] The thesis is composed of two essays on empirical finance. The first focuses on FX markets and presents measures of interest rates shortterm structure slope changes for the US and other G10 countries using 3- and 6-month futures contracts. These changes in slopes have immediate impact on currency returns but also a strong delayed effect over the following weeks, implying that currencies are predictable both in and outof-sample. Investors that condition on slope to tactically trade a long G10 portfolio improve Sharpe ratios to 0.4-0.9, relative to 0.15 for a buy-andhold strategy. A dollar-neutral currency portfolio that sorts G10 country currencies on the cross-section slope also deliver higher Sharpe ratios than other currency strategies, such as the carry trade. These findings are compatible with delayed currency market reaction to information in interest rates. The second essay proposes a novel measure that solely use crosssectional dispersion information on CAPM betas to forecast aggregate market returns for the US. This choice of predictors is based on simple theoretical arguments that measures associated with the dispersion of CAPM betas, in some settings, should be related with expected future market returns. We find that these dispersion measures do indeed forecast market risk premium over multiple horizons and deliver high in-sample and out-of-sample predictive power: out-of-sample R2 reaches up to 10 percent at the annual frequency (0.7 percent monthly) and are robust to different estimation windows. Unlike most measures in the literature, ours is not a price- or valuation-based ratio. Our approach is also an alternative to models that use the cross-section of valuation ratios to infer the conditional market risk premium. Our measures vary with the business cycle and correlate with other commonly used forecasting variable such as dividend-price or consumption-wealth ratios, but they provide explanatory power above and beyond the standard predictors. Our findings provide additional evidence that the betas dispersion across time is a function of time varying risk premium.

[pt] Esta tese é composta por dois capítulos. O primeiro capítulo mostra que a presença de efeitos lead-lag no mercado de ações dos EUA é um fenômeno mais amplo do que previamente reportado pela literatura e está associado à existência de momentum de fatores de um dia. Os efeitos lead-lag estão presentes na frequência diária, sempre que as ações são expostas ao mesmo fator de risco, difundidas por quase 100 fatores. Este fenômeno não é explicado pelo efeito por indústria, reportado previamente pela literatura, efeitos de firmas de baixo valor de mercado reagindo a firmas com maior valor de mercado, assim como outros efeitos de lead-lag. O momentum de fatores de um dia está diretamente relacionado à existência de autocovariância cruzada entre ações expostas aos mesmos fatores de risco e está presente tanto na seção transversal quanto na série temporal. O momentum do fator de um dia é rentável mesmo após os custos de negociação e não apresenta quedas bruscas como outras estratégias de momentum. O momentum do fator de um mês é absorvido pelo momentum do fator de um dia, apresentando alfa negativo. A relevância do efeito do primeiro dia é confirmada com técnicas de machine learning. As reversões de curto prazo em ações também se tornam mais fortes depois de controlarmos para esse efeito de autocovariância cruzada que vem pelo componente de fatores. O segundo capítulo mostra como o momentum de fatores impacta o desempenho das estratégias de reversão de curto prazo em ações nos Estados Unidos. Benefícios significativos no desempenho podem ser alcançados se os efeitos do momento do fator forem considerados na construção de estratégias de reversão. As estratégias tradicionais de reversão de curto prazo em ações padrão têm uma exposição negativa ao momentum de fatores, uma vez que vendem as ações vencedoras de curto prazo que, em média, estão mais expostas aos fatores vencedores de curto prazo e compram ações perdedoras de curto prazo que, em média, estão mais expostas aos fatores perdedores de curto prazo. A melhor maneira de neutralizar esse efeito que prejudica a rentabilidade da reversão de curto prazo é proteger simultaneamente as exposições das ações a um conjunto elevado de fatores de risco. Por exemplo, o hedge feito apenas para os 3 fatores Fama-French não elimina completamente a exposição ao momentum de fatores. Classificar ações pelo usando o resíduo dos retornos não é tão eficiente quanto classificar nos retornos totais, pois tal estratégia não neutraliza completamente a exposição negativa ao momentum do fator. Propomos uma estratégia de reversão totalmente hedgeada que, diferentemente das estratégias convencionais de reversão de curto prazo, é lucrativa após os custos de transação, que não apresenta quedas bruscas como outras estratégias de momentum tradicional, que tem índice de Sharpe 2,5 vezes maior do que as estratégias de reversão convencionais e que é lucrativa mesmo se for restrita a apenas a ações com alto valor de mercado. / [en] This thesis is composed by two chapters. The first chapter shows that the presence of lead-lag effects in the US equity market is a broader phenomenon than previously found in the literature and is associated with the existence of a strong one-day factor momentum. Lead-lag effects are present whenever stocks are exposed to the same common risk factor, holding for almost 100 factors on a daily frequency. This phenomenon is not explained by the previously reported industry, large-cap to small-cap and other lead-lag effects. One-day factor momentum is directly related to the existence of factor-based stock cross-autocovariance and is present both in the cross-section and the time series. One-day factor momentum is profitable after trading costs and does not present crashes. One-month factor momentum is subsumed by one-day factor momentum with negative alpha in spanning tests. The relevance of the one-day effect is confirmed with machine learning techniques. Short-term reversals in stocks also become stronger after we control for this factor-based cross-autocovariance pattern. The second chapter shows how factor momentum impacts the performance of standard short-term single-equity reversal strategies in the US equity market. Significant benefits in performance can be achieved if the effects of factor momentum is considered in the construction of reversal strategies. Standard short-term reversal strategies have a negative exposure to factor momentum since they sell winner stocks that on average are more exposed to the winner factors and buy loser stocks that on average are more exposed to loser factors. The best way to neutralize this effect that drags down short-term reversal performance is to hedge stocks exposures simultaneously to a very large set of factors. For instance, hedging only with the 3 Fama-French factors does not eliminate the exposure to factor momentum. Sorting stocks using residual returns is not as efficient as sorting on total returns as it does not completely neutralize the negative exposure to factor momentum. We propose a fully-hedged reversal strategy that, differently from conventional short-term reversal strategies, is profitable after trading costs, that do not present crashes, that has Sharpe ratio 2.5 times higher than the conventional reversal strategies and that is profitable even if we restrict our sample to only large-cap stocks.