Börshandlade fonder (ETF:er) blir alltmer populära som spar- och investeringsalternativ. Antalet ETF:er och variationen av dessa ökar stadigt. Denna studie ser på ETF:er likställda aktier (ur ett värdepappersperspektiv) och syftar till att applicera momentumstrategier på den amerikanska ETF-marknaden, likt Jegadeesh och Titman (1993), för att testa sambandet mellan ETF:ers kapitalvärde och riskjusterad överavkastning, estimerat genom Jensens alfa. Med utgångspunkt från Banz (1981) som visar på samband mellan investeringar i småbolagsaktier och högre riskjusterad avkastning (än motsvarande investeringar i stora bolags aktier). Testet har baserats på portföljer sammansatta på momentum- och contrarianstrategier för att utröna om en ETFs storlek på kapitalvärde är avgörande för en ETFs avkastning. Populationen har delats upp efter kapitalvärde i grupper om 25% största respektive 25% minsta, varefter portföljer skapats med momentumvinnare och momentumförlorare inom respektive grupp. Resultaten är mångtydiga och är i en del fall i linje med forskning som stödjer teorier om sambandet mellan lågt kapitalvärde och hög riskjusterad överavkastning. I andra fall är resultatet motsatt. Dock ger de flesta observationer inte statistisk signifikans och sambandet kan därför inte styrkas statistiskt. Resultatet tyder snarare på att ett samband mellan dessa två variabler inte föreligger.

Submitted by Vinicius Vivi (vvivo@globo.com) on 2015-12-11T16:13:49Z No. of bitstreams: 1 VINICIUS_REV_FINAL_V6.pdf: 2561407 bytes, checksum: e50f00c4d97dd15e8649ad675f7eb9e2 (MD5) / Rejected by Fabiana da Silva Segura (fabiana.segura@fgv.br), reason: Prezado Vinicius Por favor alterar onde constar o nome : Escola de Administração de Empresas de São Paulo - substituir por Escola de Economia de São Paulo Proceder com a alteração e postar novamente on 2015-12-11T16:39:19Z (GMT) / Submitted by Vinicius Vivo (vvivo@globo.com) on 2015-12-21T19:43:06Z No. of bitstreams: 1 VINICIUS_REV_FINAL_V6.pdf: 2561407 bytes, checksum: e50f00c4d97dd15e8649ad675f7eb9e2 (MD5) / Approved for entry into archive by Fabiana da Silva Segura (fabiana.segura@fgv.br) on 2015-12-22T09:56:12Z (GMT) No. of bitstreams: 1 VINICIUS_REV_FINAL_V6.pdf: 2561407 bytes, checksum: e50f00c4d97dd15e8649ad675f7eb9e2 (MD5) / Made available in DSpace on 2015-12-22T11:04:42Z (GMT). No. of bitstreams: 1 VINICIUS_REV_FINAL_V6.pdf: 2561407 bytes, checksum: e50f00c4d97dd15e8649ad675f7eb9e2 (MD5) Previous issue date: 2015-11-30 / As the financial commodities market is developing and introducing new commodity indices globally, today they are divided into three generations. In 2014, the Gross Domestic Product (GDP) of agribusiness in Brazil represented an amount of R$ 1.18 trillion. This is equivalent to 24% of the total. In addition, Brazil is one of the major powers of the world for the production and exportation of commodities and through BM&FBOVESPA, the country is growing in the financial market by actions to increase liquidity, especially in futures and options market. Against this background, and given its importance in the market context and also to the area of knowledge in which it operates, this study suggest the creation of a commodities index referenced in the Brazilian derivatives market is possible, through which, financial agents can replicate its return and to serve as a benchmark for strategic investment decisions. The data collected and analyzed confirmed the possibility for companies with financial exposure in different commodities, and in different sectors, could diversify their investment portfolios. / Como o mercado financeiro de commodities vem se desenvolvendo globalmente, foram introduzidos novos índices de commodities, hoje dividido em três gerações. Em 2014, o Produto Interno Bruto (PIB) do agronegócio no Brasil representou um montante de R$ 1,18 trilhões equivalente a 24% do total. Além disso, o Brasil está entre as maiores potências do mundo em relação à produção e exportação de commodities e através da BM&FBOVESPA, o país amadurece no mercado financeiro promovendo ações para aumento da liquidez, sobretudo em contratos futuros e de opções sobre commodities. Diante deste cenário, e dada a sua importância no contexto mercadológico e também para a área do conhecimento em que se insere o tema estudado, este sugere a criação de um índice de commodities referenciado no mercado de derivativos brasileiro, por meio do qual os agentes financeiros consigam replicar o respectivo retorno e que sirva como benchmark para decisões estratégicas de investimentos. Os dados levantados e analisados confirmaram essa possibilidade para as empresas com exposição financeira em diferentes commodities em diferentes setores, e fonte para diversificação de carteira de investimento.

La méthode de la fonctionnelle de la densité d’énergie (Energy Density Functional, EDF) est une approche phénoménologique permettant de calculer les propriétés élémentaires des noyaux atomiques (masses, dimensions et formes) a l’approximation du champ moyen. Elle permet également d’obtenir les spectres correspondant à certains états collectifs (vibrationnels et rotationnels) en mettant en œuvre des calculs ≪ au-delà ≫ du champ moyen (avec la méthode de la coordonnées génératrice (GCM) ou le mécanisme de brisure/restauration de symétries). Cette méthode, applicable sur toute la charte des noyaux hormis les plus légers, connait un regain d’intérêt avec le développement de nouveaux faisceaux radioactifs tels SPIRAL2 au GANIL. Des travaux récents ont montré qu’une description des états collectifs des noyaux nécessitait la modification et la généralisation des formes habituellement utilisées pour la fonctionnelle de la densité d’énergie, notamment avec la suppression de termes dépendants de la densité. Il a également été montre que les fonctionnelles de Skyrme pouvaient conduire à des instabilités et transitions de phases non physiques dans la matière nucléaire et les noyaux, en particulier lorsque le système est autorisé à briser des symétries. Le but de ce travail de thèse a été d’achever la construction de la fonctionnelle de Skyrme généralisée, initiée par J. Sadoudi (Constraints on the nuclear energy density functional and new possible analytical forms., CEA Saclay, 2011), puis d’en ajuster les paramètres. Le terme dépendant de la densité est remplacé par un potentiel d’interaction à 3 et 4 corps et la fonctionnelle associée est utilisée pour décrire le champ moyen et le champ d’appariement. Il a été nécessaire de construire un nouveau protocole d’ajustement afin de contraindre les paramètres de cette fonctionnelle. En particulier, l’ajustement est fait avec des contraintes sur les instabilités précédemment mentionnées. Parallèlement a la construction du protocole d’ajustement, l’utilisation de techniques d’analyse covariante a permis, dans un premier temps, d’estimer les erreurs statistiques associées a chaque paramètre ajuste et les barres d’erreurs des observables calculées. Dans un second temps, nous avons étudié les corrélations existantes entre les différentes contraintes choisies dans le protocole, ce qui a permis de mettre en évidence certains défauts de ces nouvelles interactions généralisées. Enfin, dans un travail annexe, nous avons utilisé des méthodes semiclassiques de type Thomas-Fermi modifiées (MTF) afin de dériver une formule analytique permettant d’estimer la valeur du coefficient d’énergie de surface d’une interaction de Skyrme généralisée. En réajustant les paramètres de la fonctionnelle SLy5 avec une contrainte sur ce coefficient, nous avons montré qu’il était fortement corrèle aux propriétés de fission des actinides. Nous aurons ainsi à notre disposition, grâce à cette estimation du coefficient d’énergie de surface, un outil simple à mettre en œuvre dans des protocoles d’ajustements futurs qui permettra de contraindre les propriétés de fission des noyaux / -

En el ámbito de las centrales nucleares existe un gran interés en el estudio de las características de fiabilidad, mantenibilidad y disponibilidad de sus equipos y su influencia en la seguridad y la economía de las centrales. Por todo ello, la toma de decisiones sobre la mejora y, en su caso, optimización de las pruebas y del mantenimiento en centrales nucleares se puede mejorar con la consideración simultánea de criterios RAMS+C. El proceso para encontrar los procedimientos de pruebas y mantenimientos óptimos debe equilibrar los logros alcanzados en base a los criterios RAMS+C. El reto que plantea la presente tesis es, por un lado, la necesidad de formular nuevos modelos que permitan representar de forma explícita el efecto de las pruebas y mantenimientos sobre dichos criterios RAMS+C. En segundo lugar, se requiere del estudio de la viabilidad de nuevos métodos de optimización capaces de manejar de forma adecuada tanto la complejidad de los nuevos modelos desarrollados como el gran número de variables de decisión que pueden estar involucradas en el proceso de optimización en base a múltiples criterios RAMS+C y además, que sean capaces de obtener buenos resultados en la optimización en presencia de incertidumbres asociadas tanto a parámetros, modelos o variables de decisión, y que afectan a la toma de decisiones. Para la resolución del problema de optimización multicriterio que se plantea se ha seleccionado como herramienta de optimización evolutiva los algoritmos genéticos, los cuales han mostrado una alta eficiencia en la resolución de problemas complejos. El objetivo pues de la presente tesis se centra en el desarrollo de nuevos modelos y métodos necesarios para acometer el proceso de optimización de pruebas y mantenimientos en base a criterios RAMS+C, con aplicación a sistemas de seguridad de centrales nucleares. Se han propuesto varias aproximaciones que modelan el comportamiento de los equipos con independencia temporal. / Villanueva López, JF. (2011). Optimización Evolutiva y Multiobjetivo en base a criterios RAMS+C para Centrales Nucleares [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/11446 / Palancia

Diploma thesis deals with the evaluation and the comparison of the performance of mutual funds and investment funds with a focus on the real estate sector. The essence and principles of mutual funds, ETF and REIT are presented, and the resulting weaknesses and advantages. According to the selected indicators, the profitability, riskiness and expense of the investment opportunities are examined and investment recommendations for management of an investment company and potential retail investors are established.

This thesis deals with exchange traded funds (ETFs). The theoretical part is focused on familiarization with the issue from a broader perspective in the form of collective investment characteristics, leading through the current trends in this area. This part is followed by chapter about ETF, including specific areas. Selected classic and leveraged ETFs are analyzed in the practical part for the period between 2010 and 2015. Funds are analyzed on the basis of NAV in the terms of return and risk represented by selected indicators. Results are commented in a broader context in summary and discussion chapter as well as recommendations.

In this diploma thesis we gather information on European mutual funds and ETFs that would help to inform the decision of an investment manager. We cre-ated OLS models for three types of mutual funds - money market, bond and equity – to demonstrate a relationship between funds' volatility and their annualised return. We then utilised VAR models to test Granger causation between an ETF and its tracking index using their net asset value.

This Diploma thesis is focused on performance comparison between mutual funds and classic ETF. In the theoretic part, essence and the principle of functioning of mutual funds and ETF are described. In conclusion of the theoretical part is included an overview of empirical studies that are focused on the same topic. In the empirical part is performed comparison between mutual funds and ETF which are selected from 4 economic sectors and oriented on the American market exclusively. Funds performance comparison is performed from the perspective of chosen investment time horizon and in the context of declining, growing and stagnating of the American stock market. Investment recommendation is made according to results obtained from the empirical part.

京都大学 / 0048 / 新制・課程博士 / 博士(経済学) / 甲第21326号 / 経博第584号 / 新制||経||287(附属図書館) / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 砂川 伸幸, 教授 徳賀 芳弘, 教授 澤邉 紀生 / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DFAM

根據研究指出 ETF 資產規模近幾年快速成長，元大台灣卓越 50 基金因市場 規模大等優勢受到投資人的青睞，賴以巨量資料的發展使得文字探勘技術成熟， 故本研究希冀提出一套情感分析的價格預測模型，提升投資者的報酬率。 過往學者以文章中的單詞作為文字探勘的分析單位，常會產生同義詞、多義 詞的問題，因此提出情感型態分析的監督式學習方法建立模型。另外為了解決監 督式學習難以取得訓練資料的限制，本研究混合非監督式學習方法進行主題分群 與情緒傾向標注。 本研究建立台灣股市新聞文本資料集，並篩選熱門議題詞詞庫，進行非監督 式的 LDA 主題模型，發現在 2016 年總統選舉期間，媒體對於公司相關議題的注 意力降低，使得相關的文本數量大幅減少;另外在情緒傾向標注階段，因混和了 NTUSD、知網及自行擴充演算法的情感詞庫，能夠將 10%中性詞彙產生極性判 斷、96%的文本標注情緒傾向。 視覺化工具分析結果指出，DIF-MACD 能夠預測台灣卓越 50 基金的長期走 勢，而新聞情緒指數則在短期的價格波動上表現良好，且在主題模型分群中，總 體經濟、公司維運類別的新聞情緒指數具有約 1-2 日領先指標特性，對於後續的 價格預測模型有所助益。 在監督式情感分析方法，為解決上述同義詞、多義詞的問題，本研究採用型 態分類模型於中文文本，並與向量空間模型、支援向量機等方法做比較。實驗結 果指出優化的型態分類模型，並結合台灣加權股價指數，表現相對良好，F1- Measure 可達 85%。進一步討論新聞情緒對於價格預測的重要性，發現在非交易 時間序列中的新聞情緒，能夠對 0050 的價格波動產生影響。 / The past research points out that the scale of ETF assets has been growing rapidly in recent years. Yuanta Taiwan Top 50 ETF is popular with investors because of the advantages of large market scale. Through the development of Big Data, the technology of Text Mining becomes mature. Thus, we analyze the price forecast model to raise the investors' rate of return. The research of Text Mining used to take the document term to analyze, but it often results in the problem with synonym and polysemy. Therefore, this research proposes a supervised learning method of sentiment pattern analysis. In addition, in order to solve the problem with training data about the supervised learning method, we mix the unsupervised learning method to carry out the subject grouping and sentimental tendency. In this study, we establish the news dataset and screen it as popular terms that are used to an unsupervised method of LDA model. The result points out that the number of news about company dropped significantly during the 2016 Taiwan president election because of the change of media sensation. Moreover, we create the sentiment dictionary that can determine the polarity of 10% neutral terms and the emotional tendency of 96% documents by mixing the NTUSD, HowNet knowledge Database and the self-expansion algorithm. Through the data visualization, the result shows that the curve of DIF-MACD is able to predict the long-term trend of 0050, while the sentiment index of the news makes a good showing in the short-term price volatility. Besides, the news sentiment index of the subjects that belong to general economy and company has about 1 to 2 day leading indicators. Eventually, we employ the Sentiment Pattern Taxonomy Model(PTM) in Chinese texts as supervised learning method and compare with VSM and SVM. The experiment result shows that PTM combined with Taiwan Weighted Stock Index is the best when its F1-Measure is up to 85%. Apart from this, we find that the sentiment index of the news in non-trading time can influence the price volatility of 0050.