This thesis addresses asset pricing in Chinese A-share stock markets using a dataset consisting of all shares listed in Shanghai and Shenzhen stock exchanges from January 1997 to December 2007. The empirical work is carried out based on two theoretical foundations: the efficient market hypothesis and behavioural finance. It examines and compares the validity of two traditional asset pricing models and two behavioural asset pricing models. The investigation is initially performed within a traditional asset pricing framework. The three-factor Fama-French model is estimated and then augmented by additional macroeconomic and bond market variables. The results suggest that these traditional asset pricing models fail to explain fully the time-variation of stock returns in Chinese stock markets, leaving non-normally distributed and heteroskedastic residuals, calling for further explanatory variables and suggesting the existence of a structure break. Indeed, the macroeconomic and bond market factors provide little help to the asset pricing model. Using the Fama-French model as the benchmark, further research is done by investigating investor sentiment as the third dimension beside returns and risks. Investor sentiment helps explain the mis-pricing component of returns in the Fama-French model and the time-variation in the factors themselves. Incorporating investor sentiment into the asset pricing model improves the model performance, lessening the importance of the Fama-French factors, and suggesting that in China, sentiment affects both the way in which investors judge risks as well as portfolio returns directly. The sentiment effect on asset pricing is also examined under a nonlinear Markov-switching framework. The stochastic regime-dependent model reveals that stock returns in China are driven by fundamental factors in bear and low volatility markets but are prone to sentiment and become uncoupled from fundamental risks in bull and high volatility markets.

Ingeniero Civil Indusrial / La alta penetración de Twitter en Chile ha favorecido que esta red social sea utilizada por empresas, políticos y organizaciones como un medio para obtener información adicional de las opiniones de usuarios acerca de sus productos, servicios o ellos mismos. Al ser los comentarios en Twitter, por defecto, de carácter público, se pueden analizar con el fin de extraer información accionable. En particular las empresas además de estar interesadas en la información cuantitativa, les interesa saber bajo qué polaridad se efectúan estas menciones, por cuanto una variación positiva en el número de comentarios puede deberse a un mayor número de menciones tanto positivas como negativas. Si bien existen un número considerable de softwares que vienen con la funcionalidad de detección de polaridad de sentimientos, estos no son de mucha utilidad ya que la forma en que interactúa el usuario chileno con esta plataforma está llena de modismos propios de nuestro lenguaje local y abreviaciones que se deben principalmente a la limitación de caracteres de Twitter. Al ser esta una industria inmadura en Chile, la tarea de detección de polaridad de sentimientos, se está realizando de forma manual por agencias publicitarias y otro tipo de empresas, pero dado el gran número de comentarios que se producen minuto a minuto, esta tarea resulta muy demandante en tiempo y dinero. Para resolver este tipo de problemáticas se utilizan técnicas de aprendizaje automático con el fin de entrenar un algoritmo que luego pueda determinar si un comentario es positivo, negativo o neutro, campo que se conoce como sentiment analysis. Mientras más datos sean procesados para el entrenamiento del algoritmo, mejor es el desempeño del clasificador y como en Twitter es sencillo obtener comentarios mediante su API, a diferencia de la web, se han formulado técnicas para generar automáticamente la corpora que contiene los tweets de entrenamiento para cada una de las clases y así sacar provecho de esta propiedad. En este trabajo se profundiza el uso de una metodología semiautomática basada en emoticons para la generación de una corpora de tweets para la detección de polaridad de sentimientos en Twitter. Esto se realiza introduciendo un nuevo enfoque para la consolidación de los datos de entrenamiento mediante filtros que mejoran el etiquetado automático. Esto permite prevenir la aparición de comentarios erráticos y que causan ruido en las fases de entrenamiento y clasificación. Además se introduce una nueva clase de tweets que no se había considerado anteriormente, que consiste de tweets que carecen de información suficiente para clasificarlos como positivos, negativos o neutros, por lo que clasificarlos en alguna de estas clases disminuye la precisión del sistema. Evaluaciones experimentales mostraron que el uso de esta cuarta clase denominada irrelevante con el criterio de filtros presentado para la generación de la corpora, mejora el desempeño del sistema. Además se comprobó experimentalmente que el uso de una corpora generada en base a tweets chilenos clasifican mejor a los comentarios originados por usuarios locales.

The Vector Space Model (VSM) of text representation suffers a number of limitations for text classification. Firstly, the VSM is based on the Bag-Of-Words (BOW) assumption where terms from the indexing vocabulary are treated independently of one another. However, the expressiveness of natural language means that lexically different terms often have related or even identical meanings. Thus, failure to take into account the semantic relatedness between terms means that document similarity is not properly captured in the VSM. To address this problem, semantic indexing approaches have been proposed for modelling the semantic relatedness between terms in document representations. Accordingly, in this thesis, we empirically review the impact of semantic indexing on text classification. This empirical review allows us to answer one important question: how beneficial is semantic indexing to text classification performance. We also carry out a detailed analysis of the semantic indexing process which allows us to identify reasons why semantic indexing may lead to poor text classification performance. Based on our findings, we propose a semantic indexing framework called Relevance Weighted Semantic Indexing (RWSI) that addresses the limitations identified in our analysis. RWSI uses relevance weights of terms to improve the semantic indexing of documents. A second problem with the VSM is the lack of supervision in the process of creating document representations. This arises from the fact that the VSM was originally designed for unsupervised document retrieval. An important feature of effective document representations is the ability to discriminate between relevant and non-relevant documents. For text classification, relevance information is explicitly available in the form of document class labels. Thus, more effective document vectors can be derived in a supervised manner by taking advantage of available class knowledge. Accordingly, we investigate approaches for utilising class knowledge for supervised indexing of documents. Firstly, we demonstrate how the RWSI framework can be utilised for assigning supervised weights to terms for supervised document indexing. Secondly, we present an approach called Supervised Sub-Spacing (S3) for supervised semantic indexing of documents. A further limitation of the standard VSM is that an indexing vocabulary that consists only of terms from the document collection is used for document representation. This is based on the assumption that terms alone are sufficient to model the meaning of text documents. However for certain classification tasks, terms are insufficient to adequately model the semantics needed for accurate document classification. A solution is to index documents using semantically rich concepts. Accordingly, we present an event extraction framework called Rule-Based Event Extractor (RUBEE) for identifying and utilising event information for concept-based indexing of incident reports. We also demonstrate how certain attributes of these events e.g. negation, can be taken into consideration to distinguish between documents that describe the occurrence of an event, and those that mention the non-occurrence of that event.

Dans l’apologétique de Blaise Pascal, le rôle du corps est ambigu. Source des concupiscences, le corps éloigne les hommes de la connaissance des vérités. L’homme, composé d’âme et de corps, n’a aucune similitude avec Dieu, être purement spirituel. Mais selon Pascal, c’est ce constat qui fournit à l’homme les raisons de la nécessaire croyance en Dieu, et qui fait que celle-ci exige l’existence du corps. La première en est que la foi ne se donne pas pour objet des connaissances démontrables par la raison humaine. Elle est le seul moyen d’accès à la vérité de Dieu, que la raison seule n’est pas capable d’atteindre puisqu’elle est privée de sa fonction originelle depuis que l’homme est doté de sa chair. Ensuite, la foi prend la forme de pratiques physiques, du moins dans son stade initial : l’adoption des actes d’un autre qui est déjà croyant. L’initié, pratiquant des mouvements rituels sans s’interroger sur leur sens, est persuadé de la justesse de sa foi. Enfin, d’après l’apologiste, la dévotion à Dieu permet à l’homme de jouir de l’espérance d’une autre vie ; or c’est là le suprême bonheur de la vie terrestre. L’être humain n’obtiendrait la béatitude au moment de sa mort qu’après avoir passé sa vie dans un effort sincère et continuel pour mériter d’obtenir la grâce de Dieu et dans la crainte permanente d’être délaissé de lui. En assumant ce devoir, on acquiert un bonheur incomparable, puisqu’il offre la possibilité de réaliser l’énorme gain que représente la félicité infinie et éternelle à la suite de sa brève existence. La foi, selon Pascal, c’est le chemin vers la certitude du salut, autrement dit, la béatitude en puissance. / We comment on the epistemology, ontology and axiology of the notion of man as a body or “corps” according to Blaise Pascal, in order to shed light on the concept in relation to his apologetic views. According to Pascal, “customs” and “sentiments”, the two fundamental ways of understanding the human form, provide man with secular and religious beliefs, which both allow and yet prevent him from transcending his earthly state. This equates to the ambiguous nature of realities which Pascal calls “corps”: The term refers both to purely profane matters considered as objects for scientific research, and to religious ones with their inherent symbolism, the subject of veneration. To Pascal, man, being of flesh and blood, is fated to be caught between greatness and misery; it is this axiologically ambiguous position that demands from man faith — the hope for the eternal and spiritual life, which is denied him in life on earth.

We can leverage data and complex systems science to better understand society and human nature on a population scale through language --- utilizing tools that include sentiment analysis, machine learning, and data visualization. Data-driven science and the sociotechnical systems that we use every day are enabling a transformation from hypothesis-driven, reductionist methodology to complex systems sciences. Namely, the emergence and global adoption of social media has rendered possible the real-time estimation of population-scale sentiment, with profound implications for our understanding of human behavior. Advances in computing power, natural language processing, and digitization of text now make it possible to study a culture's evolution through its texts using a "big data" lens. Given the growing assortment of sentiment measuring instruments, it is imperative to understand which aspects of sentiment dictionaries contribute to both their classification accuracy and their ability to provide richer understanding of texts. Here, we perform detailed, quantitative tests and qualitative assessments of 6 dictionary-based methods applied to 4 different corpora, and briefly examine a further 20 methods. We show that while inappropriate for sentences, dictionary-based methods are generally robust in their classification accuracy for longer texts. Most importantly they can aid understanding of texts with reliable and meaningful word shift graphs if (1) the dictionary covers a sufficiently large enough portion of a given text's lexicon when weighted by word usage frequency; and (2) words are scored on a continuous scale. Our ability to communicate relies in part upon a shared emotional experience, with stories often following distinct emotional trajectories, forming patterns that are meaningful to us. By classifying the emotional arcs for a filtered subset of 4,803 stories from Project Gutenberg's fiction collection, we find a set of six core trajectories which form the building blocks of complex narratives. We strengthen our findings by separately applying optimization, linear decomposition, supervised learning, and unsupervised learning. For each of these six core emotional arcs, we examine the closest characteristic stories in publication today and find that particular emotional arcs enjoy greater success, as measured by downloads. Within stories lie the core values of social behavior, rich with both strategies and proper protocol, which we can begin to study more broadly and systematically as a true reflection of culture. Of profound scientific interest will be the degree to which we can eventually understand the full landscape of human stories, and data driven approaches will play a crucial role. Finally, we utilize web-scale data from Twitter to study the limits of what social data can tell us about public health, mental illness, discourse around the protest movement of #BlackLivesMatter, discourse around climate change, and hidden networks. We conclude with a review of published works in complex systems that separately analyze charitable donations, the happiness of words in 10 languages, 100 years of daily temperature data across the United States, and Australian Rules Football games.

This paper proposes an audience selection framework for online brand advertising based on user activities on social media platforms. It is one of the first studies to our knowledge that develops and analyzes implicit brand-brand networks for online brand advertising. This paper makes several contributions. We first extract and analyze implicit weighted brand-brand networks, representing interactions among users and brands, from a large dataset. We examine network properties and community structures and propose a framework combining text and network analyses to find target audiences. As a part of this framework, we develop a hierarchical community detection algorithm to identify a set of brands that are closely related to a specific brand. This latter brand is referred to as the "focal brand." We also develop a global ranking algorithm to calculate brand influence and select influential brands from this set of closely related brands. This is then combined with sentiment analysis to identify target users from these selected brands. To process large-scale datasets and networks, we implement several MapReduce-based algorithms. Finally, we design a novel evaluation technique to test the effectiveness of our targeting framework. Experiments conducted with Facebook data show that our framework provides significant performance improvements in identifying target audiences for focal brands.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

The objectives of this thesis are: first, to investigate the impact of investor sentiment in UK financial markets in different investment intervals through the construction of separate sentiment measures for UK investors and UK institutional investors; second, to examine institutional herding behaviour by studying UK mutual fund data; third, to explore the causal relation between institutional herding and investor sentiment. The study uses US, German and UK financial market data and investor sentiment survey data from 1st January 1996 to 30th June 2011. The impact of investor sentiment on UK equity returns is studied both in general, and more specifically by distinguishing between tranquil and financial crisis periods. It is found that UK equity returns are significantly influenced by US individual and institutional sentiment and hardly at all by local UK investor sentiment. The sentiment contagion across borders is more pronounced in the shorter investment interval. The investigation of institutional herding behaviour is conducted by examining return dispersions and the Beta dispersions of UK mutual funds. Little evidence of herding in return is found, however strong evidence of Beta herding is presented. The study also suggests that beta herding is not caused by market fundamental and macroeconomic factors, instead, it perhaps arises from investor sentiment. This is consistent between closed-end and open-ended funds. The relation between institutional herding and investor sentiment is investigated by examining the measures of herding against the measures of investor sentiment in the UK and US. It suggests that UK institutional herding is influenced by investor sentiment, and UK institutional sentiment has a greater impact as compared to UK market sentiment. Open-end fund managers are more likely to be affected by individual investor sentiment, whereas closed-end fund managers herd on institutional sentiment.