L'objectif de cette thèse est de développer des solutions efficaces pour laclassification interactive des images de télédétection. Cet objectif a étéréalisé en répondant à quatre questions de recherche.La première question porte sur le fait que les descripteursd'images proposées dans la littérature obtiennent de bons résultats dansdiverses applications, mais beaucoup d'entre eux n'ont jamais été utilisés pour la classification des images de télédétection. Nous avons testé douzedescripteurs qui codent les propriétés spectrales et la couleur, ainsi que septdescripteurs de texture. Nous avons également proposé une méthodologie baséesur le classificateur KNN (K plus proches voisins) pour l'évaluation desdescripteurs dans le contexte de la classification. Les descripteurs Joint Auto-Correlogram (JAC),Color Bitmap, Invariant Steerable Pyramid Decomposition (SID) etQuantized Compound Change Histogram (QCCH), ont obtenu les meilleursrésultats dans les expériences de reconnaissance des plantations de café et depâturages.La deuxième question se rapporte au choix del'échelle de segmentation pour la classification d'images baséesur objets.Certaines méthodes récemment proposées exploitent des caractéristiques extraitesdes objets segmentés pour améliorer classification des images hauterésolution. Toutefois, le choix d'une bonne échelle de segmentation est unetâche difficile.Ainsi, nous avons proposé deux approches pour la classification multi-échelles fondées sur le les principes du Boosting, qui permet de combiner desclassifieurs faibles pour former un classifieur fort.La première approche, Multiscale Classifier (MSC), construit unclassifieur fort qui combine des caractéristiques extraites de plusieurséchelles de segmentation. L'autre, Hierarchical Multiscale Classifier(HMSC), exploite la topologie hiérarchique de régions segmentées afind'améliorer l'efficacité des classifications sans perte de précision parrapport au MSC. Les expériences montrent qu'il est préférable d'utiliser des plusieurs échelles plutôt qu'une seul échelle de segmentation. Nous avons également analysé et discuté la corrélation entre lesdescripteurs et des échelles de segmentation.La troisième question concerne la sélection des exemplesd'apprentissage et l'amélioration des résultats de classification basés sur lasegmentation multiéchelle. Nous avons proposé une approche pour laclassification interactive multi-échelles des images de télédétection. Ils'agit d'une stratégie d'apprentissage actif qui permet le raffinement desrésultats de classification par l'utilisateur. Les résultats des expériencesmontrent que la combinaison des échelles produit de meilleurs résultats que leschaque échelle isolément dans un processus de retour de pertinence. Par ailleurs,la méthode interactive permet d'obtenir de bons résultats avec peud'interactions de l'utilisateur. Il n'a besoin que d'une faible partie del'ensemble d'apprentissage pour construire des classificateurs qui sont aussiforts que ceux générés par une méthode supervisée qui utilise l'ensembled'apprentissage complet.La quatrième question se réfère au problème de l'extraction descaractéristiques d'un hiérarchie des régions pour la classificationmulti-échelles. Nous avons proposé une stratégie qui exploite les relationsexistantes entre les régions dans une hiérarchie. Cette approche, appelée BoW-Propagation, exploite le modèle de bag-of-visual-word pour propagerles caractéristiques entre les échelles de la hiérarchie. Nous avons égalementétendu cette idée pour propager des descripteurs globaux basés sur leshistogrammes, l'approche H-Propagation. Ces approches accélèrent leprocessus d'extraction et donnent de bons résultats par rapport à l'extractionde descripteurs globaux. / A huge effort has been made in the development of image classification systemswith the objective of creating high-quality thematic maps and to establishprecise inventories about land cover use. The peculiarities of Remote SensingImages (RSIs) combined with the traditional image classification challengesmake RSI classification a hard task. Many of the problems are related to therepresentation scale of the data, and to both the size and therepresentativeness of used training set.In this work, we addressed four research issues in order to develop effectivesolutions for interactive classification of remote sensing images.The first research issue concerns the fact that image descriptorsproposed in the literature achieve good results in various applications, butmany of them have never been used in remote sensing classification tasks.We have tested twelve descriptors that encodespectral/color properties and seven texture descriptors. We have also proposeda methodology based on the K-Nearest Neighbor (KNN) classifier for evaluationof descriptors in classification context. Experiments demonstrate that JointAuto-Correlogram (JAC), Color Bitmap, Invariant Steerable Pyramid Decomposition(SID), and Quantized Compound Change Histogram (QCCH) yield the best results incoffee and pasture recognition tasks.The second research issue refers to the problem of selecting the scaleof segmentation for object-based remote sensing classification. Recentlyproposed methods exploit features extracted from segmented objects to improvehigh-resolution image classification. However, the definition of the scale ofsegmentation is a challenging task. We have proposedtwo multiscale classification approaches based on boosting of weak classifiers.The first approach, Multiscale Classifier (MSC), builds a strongclassifier that combines features extracted from multiple scales ofsegmentation. The other, Hierarchical Multiscale Classifier (HMSC), exploits thehierarchical topology of segmented regions to improve training efficiencywithout accuracy loss when compared to the MSC. Experiments show that it isbetter to use multiple scales than use only one segmentation scale result. Wehave also analyzed and discussed about the correlation among the useddescriptors and the scales of segmentation.The third research issue concerns the selection of training examples and therefinement of classification results through multiscale segmentation. We have proposed an approach forinteractive multiscale classification of remote sensing images.It is an active learning strategy that allows the classification resultrefinement by the user along iterations. Experimentalresults show that the combination of scales produces better results thanisolated scales in a relevance feedback process. Furthermore, the interactivemethod achieves good results with few user interactions. The proposed methodneeds only a small portion of the training set to build classifiers that are asstrong as the ones generated by a supervised method that uses the whole availabletraining set.The fourth research issue refers to the problem of extracting features of ahierarchy of regions for multiscale classification. We have proposed a strategythat exploits the existing relationships among regions in a hierarchy. Thisapproach, called BoW-Propagation, exploits the bag-of-visual-word model topropagate features along multiple scales. We also extend this idea topropagate histogram-based global descriptors, the H-Propagation method. The proposedmethods speed up the feature extraction process and yield good results when compared with globallow-level extraction approaches.

Syfte: De immateriella tillgångarna är svårvärderade och diffusa tillgångar. Det har länge pågått en diskussion angående huruvida hantering av de immateriella tillgångarna påverkar värderelevansen och användbarheten av de finansiella rapporterna. Syftet med studien är att undersöka värderelevansen av de redovisade immateriella tillgångarna för kapitalmarknaden utifrån ett redovisningsperspektiv. Metod: Metoden grundas i Ohlson-modellen. Data från årsredovisningar har samlats in och sedan analyserats genom flera multipla regressioner och en korrelationsanalys.  Resultat & Slutsats:Resultatet av studien visar att det finns ett positivt signifikant samband mellan de totala immateriella tillgångarna och aktiepriset. Resultatet av denna studie visar således att de redovisade totala immateriella tillgångarna ger värderelevant information till kapitalmarknaden. Dock visar sig endast posterna FoU och patent vara värderelevanta för kapitalmarknaden när värderelevansen av de immateriella tillgångarna uppdelat i specifika poster undersöks. Förslag till fortsatt forskning:  Ett fåtal tidigare studier undersöker tillförlitligheten av de redovisade immateriella tillgångarna. Då tillförlitlighet precis som värderelevans är en av de fundamentala egenskaper redovisningsinformation bör besitta är det intressant att även undersöka hur tillförlitlig kapitalmarknaden anser att de redovisade immateriella tillgångarna är. Då tidigare forskning hävdar att det är svårt att mäta de immateriella tillgångarna på ett tillförlitligt sätt vore det relevant att undersöka om detta stämmer. Uppsatsens bidrag: Studien ger bevis för att de totala redovisade immateriella tillgångarna är värderelevanta. Det huvudsakliga praktiska bidraget med studien är att den kan ge intressenter en bättre förståelse för hur redovisningsinformation om de immateriella tillgångarna påverkar aktiepriset.

This dissertation is a study of artificial intelligence for diagnosing the condition of high voltage bushings. The techniques include neural networks, genetic algorithms, fuzzy set theory, particle swarm optimisation, multi-classifier systems, factor analysis, principal component analysis, multidimensional scaling, data-fusion techniques, automatic relevance determination and autoencoders. The classification is done using Dissolved Gas Analysis (DGA) data based on field experience together with criteria from IEEEc57.104 and IEC60599. A review of current literature showed that common methods for the diagnosis of bushings are: partial discharge, DGA, tan- (dielectric dissipation factor), water content in oil, dielectric strength of oil, acidity level (neutralisation value), visual analysis of sludge in suspension, colour of the oil, furanic content, degree of polymerisation (DP), strength of the insulating paper, interfacial tension or oxygen content tests. All the methods have limitations in terms of time and accuracy in decision making. The fact that making decisions using each of these methods individually is highly subjective, also the huge size of the data base of historical data, as well as the loss of skills due to retirement of experienced technical staff, highlights the need for an automated diagnosis tool that integrates information from the many sensors and recalls the historical decisions and learns from new information. Three classifiers that are compared in this analysis are radial basis functions (RBF), multiple layer perceptrons (MLP) and support vector machines (SVM). In this work 60699 bushings were classified based on ten criteria. Classification was done based on a majority vote. The work proposes the application of neural networks with particle swarm optimisation (PSO) and genetic algorithms (GA) to compensate for missing data in classifying high voltage bushings. The work also proposes the application of fuzzy set theory (FST) to diagnose the condition of high voltage bushings. The relevance and redundancy detection methods were able to prune the redundant measured variables and accurately diagnose the condition of the bushing with fewer variables. Experimental results from bushings that were evaluated in the field verified the simulations. The results of this work can help to develop real-time monitoring and decision making tools that combine information from chemical, electrical and mechanical measurements taken from bushings.

This study explored how personal strivings constructs of Goal Conflict and Complementarity and primary appraisal dimensions of Motivational Relevance and Congruence (separately and in combination), related across time to immediate and long-term effects of a stressful transaction. The study was located within the transactional model of stress and integrated aspects of motivational theory, focusing on the theoretical position that within the stressful transaction the relationship between motivational factors and the individual response to an event is mediated by cognitive processes, including appraisal. Advances in transactional theory highlight the role of motivational factors (such as personal strivings) as linked to primary appraisal in the form of Motivational Relevance and Congruence. The utility of personal strivings in exploring the role of motivational factors in the stress process were highlighted. Despite an increasing theoretical focus on motivation and appraisal, research in the area is limited. Data was collected for the study through the administration of questionnaires to university students (N=152) prior to (time 1) and into (time 2) an examination period. The questionnaires used a range of self-report measures. Correlations, partial correlations and ANOVAs were used to analyze the data. The findings indicated that Goal Complementarity and Conflict directly influenced primary appraisal processes but not affective and wellbeing outcomes. It was proposed that primary appraisal processes were the conduit through which the impact of Goal Complementarity and Conflict were expressed within the stressful transaction. The results also suggested the impact of anticipatory Motivational Relevance which seemed to imply a highly “loaded” event with negative affective and long-term consequences into the event. Subjects entering the event wit h an “optimistic” demeanour indicated by high anticipatory Congruence and positive affect had increased Congruence into the event with consequent amplified positive emotions and dampened negative effects. Subjects with high anticipatory Relevance and low Congruence across the event had relatively higher scores on negative outcomes. High anticipatory Relevance and Congruence was associated with negative immediate and long-term outcomes into the event. High Relevance Congruence was generally associated with a strong emotional response, which also elicited strong positive emotion as the event unfolded. Subjects with low Relevance did not seem to hold as strong an investment in the event and reported reduced emotions and symptomology. These findings were discussed in relation to the stress, appraisal and motivation literature and their limitations and implications were explored.

Implicatura escalar é uma interpretação upper-bound (limite máximo de informatividade) de um termo escalar fraco. E.g. \'alguns\', o termo fraco da escala semântica <todos, alguns>, pode ter interpretação lower-bound/literal (\"pelo menos alguns e talvez todos\") ou upper-bound/pragmática, de implicatura escalar (\"apenas alguns mas não todos\"). Na primeira parte deste trabalho, investigamos, por vias teóricas e experimentais, a capacidade infantil de interpretação de implicaturas escalares, que entendemos, por influência da Teoria da Relevância, como a negação implicada por um termo fraco de uma pergunta explícita ou contextualmente implícita acerca do termo forte. Nossos resultados indicam domínio de implicatura escalar pelas crianças e sensibilidade capaz de flexibilizar a interpretação dos termos fracos de acordo com contextos lower-bound ou upper-bound. Assim, identificamos na já bem desenvolvida capacidade infantil de inferir intenções comunicativas alheias (mindreading) a função cognitiva central para o sucesso de sua interpretação pragmática. Dedicamos a segunda parte do texto a uma mais abrangente exploração do papel da leitura de intenções na aquisição de linguagem, ainda comprometidos com os paradigmas conceituais da Teoria da Relevância. Ultrapassando o escopo da interpretação pragmática, procuramos reunir evidências favoráveis à adoção da capacidade humana de leitura de intenções como a operação cognitiva motivadora e coesiva do processo de amadurecimento linguístico, de modo que, argumentamos, nenhuma teoria que se proponha a explicar a aquisição de linguagem deve negligenciar em seu modelo a esfera sócio-pragmática da comunicação verbal. / Scalar implicatures happen when an upper-bound interpretation (upper limit of informativeness) is given to a weak scalar term. E.g. \'some\' (the weak term of the scale <all, some>) may receive a lower-bound/literal interpretation (\"at least some and perhaps all\") or an upper-bound/pragmatic scalar implicature interpretation (\"only some but not all\"). In the first part of the work, we investigate children\'s capacity of scalar implicature interpretation through theoretical and experimental research. By influence of Relevance Theory, we assume that such interpretation must infer the negation implicated by the weak term of an explicit or contextually implicit question about the strong term. Our results indicate children\'s mastery of scalar implicature and their sensitivity to adaptation of weak term\'s interpretation according to lower-bound or upper-bound contexts. Thus, we see in the well-developed child capacity of reading other\'s communicative intentions (mindreading) the central cognitive function of their successful pragmatic interpretation. The second part is dedicated to a more embracing analysis of mindreading\'s role in language acquisition, still taking Relevance Theory\'s concepts as references. Beyond the scope of pragmatic interpretation, we sought to gather evidences that support the hypothesis of the human capacity of intentions reading as the cognitive operation that motivates language development and provides cohesion to the parts of the process. Being so, we argue that no theory that intends to explain language acquisition should neglect in its model the social-pragmatics level of verbal communication.

O setor bancário é um dos mais importantes na sociedade, pois suas atividades relacionadas ao processo de pagamentos e sua atuação como intermediário financeiro são essenciais no desenvolvimento da economia. No entanto, em razão da natureza de suas operações e dos riscos envolvidos nas mesmas, o setor bancário também é um dos que apresentam regulação mais rígida. Bancos que atuam globalmente recebem influência de regulação em esfera internacional, como por exemplo, o Acordo de Basileia, que versa sobre requerimentos mínimos de capital. A regulação bancária e, por conseguinte, o Acordo de Basileia, podem impactar as atividades dos bancos, sua captação no mercado, suas estratégias e competição. Tal impacto deve ser avaliado tanto pelos próprios bancos, como pelos reguladores, pelo mercado e pela academia. A presente pesquisa avalia o impacto da regulação bancária, em especial o Acordo de Basileia, no mercado de capitais por meio de um estudo de value relevance. Utilizando o modelo de regressão do preço (Price Regression Model - PRM) com dados de bancos dos 27 países membros do Comitê de Basileia, foi avaliado se o índice de Basileia afeta o preço das ações dos bancos. Conclui-se que o índice de Basileia é value relevant para a amostra completa. No entanto, com a retirada dos Estados Unidos da amostra, devido ao alto percentual de participação (51%) na mesma, o índice de Basileia deixou de ser value relevant. Além disso, os resultados mostraram que o índice de Basileia não é value relevant para países considerados code law (amostra completa) e nem para países adotantes do IFRS (amostra completa). / The banking sector is one of the most important in society, because its activities related to the payment process and its role as financial intermediary are essential in the development of the economy. However, because of the nature of its operations and the risks involved in them, the banking sector is also one of the most rigid. And banks that operate globally receive regulatory influence in the international arena, such as the Basle Accord, which deals with minimum capital requirements. Banking regulation, and therefore the Basle Accord, can impact the activities of banks, their capture in the market, their strategies and competition. Such an impact must be assessed by the banks themselves, regulators, the market and academy. The present research evaluates the impact of banking regulation, especially the Basel Accord, on the capital market through a value relevance study. Using the Price Regression Model (PRM) with data from banks of the 27 member countries of the Basel Committee, it was assessed whether the Basel index affects the stock price of banks. We conclude that the Basel index is value relevant for the whole sample. However, with the withdrawal of the United States from the sample, due to the high percentage of participation (51%) in the sample, the Basel index is no longer value relevant. In addition, the results showed that the Basel index is not value relevant for countries considered code law and for adopters of the IFRS (full sample).

Esta tese apresenta um modelo de relevância de opinião genérico e independente de domínio para usuários de Redes Sociais. O Social Opinion Relevance Model (SORM) é capaz de estimar a relevância de uma opinião com base em doze parâmetros distintos. Comparado com outros modelos, a principal característica que distingue o SORM é a sua capacidade para fornecer resultados personalizados de relevância de uma opinião, de acordo com o perfil da pessoa para a qual ela está sendo estimada. Devido à falta de corpus de relevância de opiniões capazes de testar corretamente o SORM, fez-se necessária a criação de um novo corpus chamado Social Opinion Relevance Corpus (SORC). Usando o SORC, foram realizados experimentos no domínio de jogos eletrônicos que ilustram a importância da personalização da relevância para alcançar melhores resultados, baseados em métricas típicas de Recuperação de Informação. Também foi realizado um teste de significância estatística que reforça e confirma as vantagens que o SORM oferece. / This thesis presents a generic and domain independent opinion relevance model for Social Network users. The Social Opinion Relevance Model (SORM) is able to estimate an opinions relevance based on twelve different parameters. Compared to other models, SORMs main distinction is its ability to provide customized results, according to whom the opinion relevance is being estimated for. Due to the lack of opinion relevance corpora that are able to properly test our model, we have created a new one called Social Opinion Relevance Corpus (SORC). Using SORC, we carried out some experiments on the Electronic Games domain that illustrate the importance of customizing opinion relevance in order to achieve better results, based on typical Information Retrieval metrics, such as NDCG, QMeasure and MAP. We also performed a statistical significance test that reinforces and corroborates the advantages that SORM offers.

O presente trabalho possui o objetivo de investigar se a adoção das IFRS mudou a relação da variação do valor justo dos derivativos no valor de mercado em instituições financeiras do mundo e em particular no Brasil entre os anos de 2005 e 2015. A abordagem desta pesquisa é empírico-analítica e para a análise dos modelos propostos foi utilizada a técnica de dados em painel e realizados testes de adequação para melhor estimação. A hipótese investigada é que a variação do valor justo dos derivativos, associada à adoção das IFRS, podem influenciar o valor de mercado da empresa. Adicionalmente, foi realizada análise complementar segregando o valor justo dos derivativos entre os que foram designados para hedge e os não de hedge. Nos modelos completos foram considerados controles como o Tamanho do banco, Índice de Basileia, a ocorrência da Crise em 2008, o número de analistas que seguem a instituição e o sistema jurídico para a amostra mundial. Após os testes empíricos, não foram encontradas relações estatisticamente relevantes entre a variação do valor justo dos derivativos após a adoção das IFRS e o valor de mercado das companhias investigadas tanto no Brasil como no mundo, de uma forma geral. A pouca liquidez do mercado de capitais do Brasil, o tratamento dos derivativos pelas normas locais do Banco Central antecedente ao IFRS e a utilização de bases de dados secundárias com grande presença de dados faltantes para a amostra mundial podem explicar esse resultado. / The objective of this study is to investigate whether the adoption of IFRS has changed the relationship of the derivatives fair value on the market value of financial institutions in the world and in particular in Brazil between 2005 and 2015. The approach of this research is empirical-analytical and for the analysis of the proposed models, the panel data technique was used and adequacy tests were performed for better estimation. The investigated hypothesis is that the variation of the derivatives fair value, associated with the adoption of IFRS, can influence the market value of the company. In addition, a complementary analysis was performed segregating the derivatives fair value between those that were designated for hedge and the non hedge. In the full models, controls such as Bank Size, Basle Index, the occurrence of the Crisis in 2008, the number of analysts who followed the institution and the legal system for the world sample were considered. After the empirical tests, there were no statistically significant relationship between the variation of the derivatives fair value after the adoption of the IFRS and the market value of the companies investigated in Brazil as in the world, in general. The low liquidity of the Brazilian capital market, the treatment of derivatives by the local standards of the Central Bank of Brazil prior to IFRS and the use of secondary databases with a large presence of missing data for the world sample may explain this result.

Att självkritiskt ställa frågan: ”Gör vi saker rätt eller gör vi rätt saker?” är vitalt om man avser förbättra det man gör. Svensk officersutbildning har bedrivits på Militärhögskolan Karlberg sedan 1792 och dess historia har satt sina spår i dagens officersutbildningssystem. Omkastningar i utbildningsupplägget är ofta en effekt av förändrat omvärldsläge och när större förändringar av utbildningssystemet sker blir också vikten av denna självkritiska fråga som störst. Skall man då fortsätta hävda att det var bättre förr eller behöver man acceptera de förändringar som sker och tänka nytt?Vår studie lyfter på denna heta sten med ett avstamp i verkligheten i form av före detta kadetters upplevelser, erfarenheter och perspektiv. Vi undersöker om det finns substans bakom den upplevda problematik mellan förväntningar och utbildning som chefen för programledningen vid Officersprogrammet ger uttryck för, i en artikel på Försvarshögskolans hemsida där han säger att: "Idag finns ibland ett glapp mellan förbandens förväntningar på de nyutbildade officerarna och den utbildning som bedrivs av Försvarshögskolan på uppdrag av regeringen. (Brorson. 2018)”Empirin är inhämtad genom intervjuer med före detta flygvapenkadetter från Officersprogrammet med ett antal års tjänstgöring i Försvarsmakten. Genom vald arbetsmetodik har datat analyserats och kodats till tre olika teman. Utifrån dessa teman har vi tagit fram en teoretisk referensram som vi granskat och utvecklat våra initiala analyser med i syfte att besvara forskningsfrågan:Hur upplever före detta kadetter att utbildningen vid Officersprogrammet förberett dem för en anställning i Flygvapnet? Våra sammanställda slutsatser pekar på ett flertal anledningar till varför Officersprogrammet inte lever upp till förväntningarna för hur utbildningen borde förbereda kadetterna inför en anställning i Flygvapnet. Vi beskriver i studien orsaker, samt ger förslag på åtgärder som Försvarsmakten och Försvarshögskolan kan vidta för att motverka denna problematik. / ”Do we do things right or do we do the right things?”  This is the self-critical question that one has to ask oneself if one aims to improve what one is doing.The Swedish Armed Forces officer training has been conducted at the Military Academy at Karlberg since 1792, and its long history has set its mark on today's educational system.  Changes in the training set-up are often a result of changes in the surrounding world.  When larger reversals in the educational system occur, as they do today, the importance of the self-critical question becomes all the greater.  Should one continue to claim that things used to be better or should one accept the spirit of change and think in new terms?Our study highlights this burning issue with a visit to reality in the form of the experiences and perspectives of former cadets.  We examine whether there is any substance behind the perceived conflict between expectations and education which the Head of the program management of the Officers Program describes in an article on the Swedish Defense College website, where he states:  ”Today, there is sometimes a gap between the expectations of the newly trained officers among the Swedish Armed Forces units and the education conducted by the Swedish National Defense University on behalf of the government.”  (Brorson. 2018)The empirical data is obtained through interviews with former Air Force cadets from the Officer Program with a number of years’ service in the Armed Forces. Through our selected work methodology, the data has been analyzed and coded into three different themes. Based on these themes, we have developed a theoretical frame of reference with which we examined and expanded our initial analysis with the purpose of providing an answer to our research question:How do former cadets experience that the education at the Officer's Program prepared them for an employment in the Air Force?Our compiled conclusions point to a number of reasons why the Officer Program does not live up to expectations concerning how the education should prepare cadets for a position in the Air Force.  We explain in the study why this occurs, and recommend measures the Swedish Armed Forces and the Swedish Defense College could implement to counteract this problem.

PINTO, K. E. F. Responsabilidade Socioambiental Corporativa: um estudo em subsidiárias brasileiras. 2011. 119 f. Dissertação (Mestrado) - Faculdade de Economia, Administração e Contabilidade, Universidade de São Paulo, São Paulo, 2011. Aborda-se nesta dissertação a relação entre subsidiárias de corporações multinacionais estrangeiras no Brasil e Responsabilidade Socioambiental Corporativa (RSC). Inicialmente caracteriza-se o surgimento de corporações multinacionais e a crescente preocupação com questões socioambientais. Fundamenta-se o estudo, basicamente, nas literaturas referentes a multinacionais, matrizes, subsidiárias, gestão socioambiental e RSC. Apresenta-se, num primeiro momento, a identificação de RSC nas subsidiárias para, depois, apresentar a relação entre RSC e outros aspectos organizacionais como autonomia, integração e relevância estratégica. Além de realizar essas análises, estuda-se também a possível relação entre a RSC da subsidiária e o nível do desenvolvimento do país-sede da matriz. Para a realização do estudo, enviou-se um questionário eletrônico para os presidentes e vice-presidentes das 1200 maiores subsidiárias por faturamento e obteve-se 172 respostas. Conclui-se, por meio de análise multivariada, que a RSC encontra-se positivamente relacionada à integração entre a matriz e sua subsidiária e também à relevância estratégica atribuída à subsidiária pela matriz. Ou seja, quanto maior a integração com a matriz e quanto mais relevante estrategicamente, mais a subsidiária desenvolve RSC. Entretanto, a RSC encontra-se negativamente relacionada à autonomia da subsidiária e ao desenvolvimento do país-sede da matriz, o que resulta nas afirmações: (i) quanto mais autônoma, menos a subsidiária desenvolve RSC; e (ii) o desenvolvimento do país-sede não influi no fato de a subsidiária possuir ou não RSC. / PINTO, K. E. F. Corporate Social Responsibility: a study in Brazilian subsidiaries. 2011. 119 f. Dissertação (Mestrado) - Faculdade de Economia, Administração e Contabilidade, Universidade de São Paulo, São Paulo, 2011. In this dissertation the relationship between subsidiaries of foreign multinational corporations in Brazil and Corporate Social Responsibility (CSR) is analyzed. Herein the emergence of multinational corporations is characterized, as well as the actual concern about social and environmental issues. The study is basically based on the literatures about multinational corporations, headquarters, subsidiaries, social and environmental management and CSR. Firstly, the identification of CSR in subsidiaries is showed; and, secondly, its relation to organizational aspects such as autonomy, integration between headquarters and their subsidiaries and subsidiary strategic relevance. Besides these analyses, the possible relationship between CSR and the development of the headquarters\' country of origin is also investigated. To accomplish this study, an electronic survey was sent to presidents and VPs of the 1200 largest subsidiaries in Brazil in terms of revenue; 172 answers were received. One may conclude, due to a multivariate analysis, that CSR is positively related to the headquarters-subsidiary integration and to subsidiary strategic relevance. It indicates, therefore, that the better integrated the subsidiary is to its headquarters and the more relevant the subsidiary becomes, then the more CSR the subsidiary develops. On the other hand, CSR is negatively related to subsidiary autonomy and to the development of the headquarters\' country of origin, which leads to the final statements: (i) the more autonomous it becomes, the less the subsidiary develops CSR; (ii) the development of the headquarters\' country of origin does not influence the emergence of CSR in the subsidiary.