The environment became an established issue on the international political agenda after World War 2. How to preserve natural resources became an issue that concerned several powerful participants, affected by changed conditions in the environmental legislation. This paper is a study of the growing environmental work within the multinational Swedish corporations Boliden, IKEA and SAS and how it was affected by the globalization. The material studied is environmentally related information produced by the companies and also interviews with the Environmental Managers. The information provided has been analyzed and compared with well-known opinions of globalization, trying to find out how the process of globalization has affected the handling of environmental work within the companies. The conclusion is the globalization brings several disadvantages, but it also results in an international legislation and co-operation between different participants, something that does have a positive effect on the environment.

Nas diversas áreas da ciência e na Tecnologia da Qualidade são realizados experimentos para estudar a influência de diversos fatores sobre características (tais como "severidade do desgaste" ou "qualidade de uma solda"), que são registradas como variáveis categóricas ordenadas. Nessa dissertação são estudados aspectos da análise de experimentos com resposta ordinal, como ferramenta da Tecnologia da Qualidade. Os métodos tradicionais de análise estatística (teste Qui-quadrado e testes não-paramétricos, por exemplo) não são eficientes nessa situação, sendo necessários métodos mais elaborados para extrair e interpretar as informações geradas nesses experimentos. Diante disso, foram estudados em detalhe dois métodos alternativos: o primeiro é a técnica da Análise de Acumulação (AA) introduzida por Taguchi (I 987) e, o outro método, baseia-se no modelo de Odds Proporcionais proposto por McCullagh (1980), através do qual é possível estimar e interpretar os efeitos dos fatores. Uma característica especial dessa resposta ordinal é a possibilidade de estar associada a uma variável contínua latente. Assim, os dados observados podem ser vistos como uma categorização dessa variável contínua, não diretamente observável. Essas técnicas de análise são ilustradas para o caso unifatorial e multifatorial, mediante dois conjuntos de dados apresentados na literatura: um estudo observacional e outro experimental. O ajuste dos modelos foi realizado através do procedimento LOGISTIC do pacote estatístico SAS e pelo software ST ATA. Com base nos resultados obtidos, verifica-se a necessidade de modelos paramétricas para estimar magnitude e direção dos efeitos que os fatores provocam na característica de qualidade, uma vez que os métodos tradicionais e a AA não extraem informação suficiente dos dados. Nesse sentido, o modelo de Odds Proporcionais parece um método simples, adequado e eficiente para a análise de dados com resposta ordinal. / In several areas o f Science and Quality o f Technology, experiments are conducted to study the influence of severa! factors on characteristics (as "worn-out severity, or "weld quality"), that are recorded as ordered categorical variables. In this thesis, some aspects o f the analysis o f experiments with ordinal response are studied as a tool of Quality of Technology. The traditional methods of statistical analysis (Chísquare test and non-parametric tests, for example) are not efficient in thís situation, and more sophisticated methods are needed to extract and to interpret the information produced in these experiments. So, two alternative methods are studied in detail: the first is the Accumulation Analysis (AA) technique introduced by Taguchí (1987). The other method is based on the Proportional Odds model proposed by McCullagh (1980), whích allows to estimate and to interpret the effects of the factors. A special characteristics of this response is that it might be associated to a continuous Iatent variable. Thus, the observed data may be seen as a categorization o f this continuous variable not directly observable. These techniques of analysis are illustrated for the single-factor and the multifactor cases, by using two data set presented in the literature: an observational study and an experimental study. The models were adjusted by the procedure LOGISTIC o f the software SAS and by the software ST ATA. The results obtained point to the need of parametric models to estimate the magnitude and direction of the effects caused by the factors in the characteristic of quality, since the traditional methods and the AA do not extract enough information of the data. In thís sense, the Proportional Odds model seems a simple, adequate and efficient method for analyzing data with ordinal response.

Em estudos epidemiológicos. Os desfechos são freqüentemente medidos em escalas ordinais, entretanto são analisados por técnicas tradicionais. como a reg ressão logística que requer desfecho binário. Consequentemente. informações importantes podem ser perdidas devido a arbitrariedade na escolha do ponto de corte. Modelos de regressão para respostas ordinais têm sido desenvolvidos e têm vantagens de reter a ordinalidade dos dados e fornecer estimativas interpretáveis para os epidemiologistas. Esse artigo descreve o modelo de odds proporcionais proposto por McGullagh e o aplica aos dados da coorte de cnanças nascidas em Pelotas em 1993. usando o pacote estatístico SAS. O objetivo desse artigo foi comparar os resultados do modelo de odds proporcionais e da regressão logística tradicional usando diferentes pontos de corte. O desfecho foi déficit de estatura para idade expressa em escore-z. com três níveis ordinais. Concluiu-se que o modelo de odds proporcionais produz uma estimativa interpretável. similar a uma razão de odds. que sumariza o efeito sobre todos pontos de corte. Na presente análise. as estimativas pontuais resultantes da regressão ordinal foram similares ás obtidas através da regressão logística tradicional usando o ponto de corte do escore-z em -1. Entretanto, a regressão ordinal produziu intervalos de confiança menores (e, portanto, aumento do poder estatístico), evitando a arbitrariedade de ter um único ponto de corte. Uma limitação para uso mais amplo do modelo ordinal é a falta de pacotes estatísticos que o incluam. / Outcomes in epidemiological studies are often measured using ordinal scales, but are nevertheless analyzed through traditional techniques such as logistic regression that require a single dychotomous outcome. Therefore, important information may be lost due to arbitrariness in the choice of a cut-off point. Regression models for ordinal responses have been developed and have the advantages of nr.aking full use of the ordinality of the data, while still providing interpretable estimates for epidemiologists. This paper describes the proportional odds model proposed by McGulla~h and applies it to data from a birth cohort in Southern Brazil. using the statistical package SAS. The objective of this paper was to compare the results of the proportional odds model and of traditional logistic regression using different cut-offs. The outcome was heighUage deficit. expressed in z-scores, with three ordinal leveis. We conclude that the proportional odds model produces an interpretable estimate. similar to an odds ratio, that summarizes the effect over ali leveis of the outcome In the present analysis, the point estimates resulting from ordinal regression were similar to those obtained through traditional logistic regression using a cut-off of -1 z-score. However, ordinal regression lead to smaller confidence intervals (and therefore increased study power), avoiding the arbitrariness of having a single cut-off point. A restriction to the wider use of the ordinal model is the lack of statistical packages that include it.

The radio spectrum dependent applications are facing a huge scarcity of the resource. To address this issue, future wireless systems require new wireless network architectures and new approaches to spectrum management. Spectrum sharing has emerged as a promising solution to address the radio frequency (RF) spectrum bottleneck. Although spectrum sharing is intended to provide flexible use of the spectrum, the architecture of the existing approaches, such as TV White Space [1] and Citizen Broadband Radio Services (CBRS) [2], have a relatively fixed sharing framework. This fixed structure limits the applicability of the architecture to other bands where the relationship between various new users and different types of legacy users co-exist. Specifically, an important aspect of sharing that has not been explored enough is the cooperation between the resource owner and the opportunistic user. Also in a shared spectrum system, the users do not have any information about the availability and duration of the available spectrum opportunities. This lack of understanding about the shared spectrum leads the research community to explore a number of core spectrum sharing tasks, such as opportunity detection, dynamic opportunity scheduling, and interference protection for the primary users, etc. This report proposes a Primary-Secondary Cooperation Framework to provide flexibility to all the involved parties in terms of choosing the level of cooperation that allow them to satisfy different objective priorities. The cooperation framework allows exchange of a probabilistic assurance: Spectrum Opportunity Duration Assurance (SODA) between the primary and secondary operations to improve the overall spectrum sharing experience for both the parties. This capability will give the spectrum sharing architectures new flexibility to handle evolutions in technologies, regulations, and the requirements of new bands being transitioned from fixed to share usage. In this dissertation we first look into the regulatory aspect of spectrum sharing. We analyze the Federal Communications Commission's (FCC) initiatives with regards to the commercial use of the 150 MHz spectrum block in the 3.5 GHz band. This analysis results into a Spectrum Access System (SAS) architecture and list of required functionalities. Then we address the nature of primary-secondary cooperation in spectrum sharing and propose to generate probabilistic assurances for spectrum opportunities. We use the generated assurance to observe the impact of cooperation from the perspective of spectrum sharing system management. We propose to incorporate primary user cooperation in the auctioning and resource allocation procedures to manage spectrum opportunities. We also analyze the improvement in spectrum sharing experience from the perspective of the primary and secondary users as a result of cooperation. We propose interference avoidance schemes that involve cooperation to improve the achievable quality of service. Primary-secondary cooperation has the potential to significantly influence the mechanism and outcomes of the spectrum sharing systems. Both the primary and secondary operations can benefit from cooperation in a sharing scenario. Based on the priorities of the primary and secondary operations, the users may decide on the level of cooperation that they are willing to participate. Also access to information about the availability and usability of the spectrum opportunity will result in efficient spectrum opportunity management and improved sharing performance for both the primary and secondary users. Thus offering assurances about the availability and duration of spectrum opportunity through primary-secondary cooperation will significantly improve the overall spectrum sharing experience. The research reported in this dissertation is expected to provide a fundamental analytical framework for characterizing and quantifying the implications of primary-secondary cooperation in a spectrum sharing context. It analyzes the technical challenges in modeling different level of cooperation and their impact on the spectrum sharing experience. We hope that this dissertation will establish the fundamentals of the spectrum sharing to allow the involved parties to participate in sharing mechanisms that is suitable to their objective priorities. / PHD / As the world of technology steps into the era of ubiquitous communication to anything and everything, a system's ability to wirelessly communicate in a heterogeneous environment plays a significant role in shaping our ways of life. The wireless communication systems and standards are evolving at an unprecedented rate to cope up with the explosive growth for uninterrupted mobile broadband service demand and the increasing diversity of high quality of service (QoS) use cases ranging from social communication and professional networking to cyber security and public safety. The rapid evolution of wireless communication systems and service applications has resulted in high demand for new and dedicated spectrum blocks in both the licensed and unlicensed bands. Also the predicted future wireless systems and applications indicate important characteristics of future broadband traffic demand: nomadic and sporadic bursty demand. But the existing static spectrum assignment limits the potential of the radio frequency spectrum resource. It imposes the challenge of spectrum scarcity onto radio spectrum dependent applications and technologies. This unprecedented increase in mobile data traffic along with the nomadic and sporadic bursts in data demand will disruptively shape the spectrum usage philosophy of the future wireless communication networks. It calls for new wireless network architectures and new approaches to spectrum management. Spectrum sharing has emerged as a promising solution to address the radio frequency (RF) spectrum bottleneck. Although spectrum sharing is intended to provide flexible use of the spectrum, the architecture of the existing approaches have a relatively fixed structure in the mechanism for which spectrum is shared. This fixed structure limits the applicability of the architecture to other bands where the relationship between various new users and different types of legacy users co-exist. Specifically, an important aspect of sharing that has not been explored enough is the cooperation between the resource owner and the opportunistic user. Also in a shared spectrum system, the users do not have any information about the availability and duration of the available spectrum opportunities. This lack of understanding about the shared spectrum leads the research community to explore a number of core spectrum sharing tasks, such as opportunity detection, dynamic opportunity scheduling, and interference protection for the primary users, etc. In this dissertation we propose a Primary-Secondary Cooperation Framework that provides flexibility to all the involved parties in terms of choosing the level of cooperation and allow them to satisfy different objective priorities. The cooperation framework allows exchange of a probabilistic assurance: Spectrum Opportunity Duration Assurance (SODA) between the primary and secondary operations to improve the overall spectrum sharing experience for both the parties. This capability will give the spectrum sharing architectures new flexibility to handle evolutions in technologies, regulations, and the requirements of new bands being transitioned from fixed to share usage. Based on their operational priorities, the users may decide on the level of cooperation that they are willing to participate. Also access to information about the availability and usability of the spectrum opportunity influences the mechanism and outcomes of the spectrum sharing systems to benefit both the Primary and Secondary users. Thus offering assurances about the availability and duration of spectrum opportunity through primary-secondary cooperation will significantly improve the overall spectrum sharing experience. The research reported in this dissertation is expected to provide a fundamental analytical framework for characterizing and quantifying the implications of primary-secondary cooperation in a spectrum sharing context. It analyzes the technical challenges in modeling different level of cooperation and their impact on the spectrum sharing experience. We hope that this dissertation will establish the fundamentals of the spectrum sharing to allow the involved parties to participate in sharing mechanisms that is suitable to their objective priorities.

Thesis (MComm (Statistics and Actuarial Science))--University of Stellenbosch, 2005. / Financial institutions invest in financial securities like equities, options and government bonds. Two measures, namely return and risk, are associated with each investment position. Return is a measure of the profit or loss of the investment, whilst risk is defined as the uncertainty about return. A financial institution that holds a portfolio of securities is exposed to different types of risk. The most well-known types are market, credit, liquidity, operational and legal risk. An institution has the need to quantify for each type of risk, the extent of its exposure. Currently, standard risk measures that aim to quantify risk only exist for market and credit risk. Extensive calculations are usually required to obtain values for risk measures. The investments positions that form the portfolio, as well as the market information that are used in the risk measure calculations, change during each trading day. Hence, the financial institution needs a business tool that has the ability to calculate various standard risk measures for dynamic market and position data at the end of each trading day. SAS Risk Dimensions is a software package that provides a solution to the calculation problem. A risk management system is created with this package and is used to calculate all the relevant risk measures on a daily basis. The purpose of this document is to explain and illustrate all the steps that should be followed to create a suitable risk management system with SAS Risk Dimensions.

本論文之研究目的是在探討今日電子、紡織、機械三種製造業所採用的製造策略類型 和其所表現的財務績效之間的關係，並深入研究其原因。 本論文所參考的文獻總共約有三十篇（含國內文獻和國外文獻），依其內容可以分為 三部份。第一部份探討各種製造策略的型態；第二部份探討各種製造策略型態和其績 效的關係；第三部份則探討目前我國電子、紡織、機械三種產業之現狀。 研究所使用的方法是先以問卷隨機抽樣調查上述三種產業的廠商，以得知他們在製造 方面的作法，所抽取的樣本計有電子業２１３家，回收３４家，機械３２０家，回收 ２８家，紡織３３０家，回收３６家。其次再使用集群分析法，將樣本以各項製造決 策為依據，區分為若干策略群。最後，將每一策略群視為某一母體的代表，以變異數 分析法，檢定各策略在各項財務績效上之表現有無顯著差異。所使用的套裝軟體有Ｓ ＰＳＳ＋（使用其中的集群分析法）和ＳＡＳ（使用其中的變異數分析法）。 所選用的自變數是各項製造策略變數，包括製程設備、產能、廠址、產品、垂直整合 、組織、供應商管理、存貨、員工領導與激勵、生產控制、品管、和生產主管角色。 自變數包括銷貨成長率、投資報酬率、利潤率、固定資產報酬率、產品品質。 根據回收樣本的資料顯示，以電子業為例，電子業皆製造策略大約可以分為四群（將 其命名為第一、第二、第三、第四策略群），這四種製造策略在銷貨額成長率和利潤 率這兩項財務績效上的表現有顯著的差異，在銷貨額成長率方面，第四群的表現最佳 ，第一群最差；在利潤率方面，則以第二群最佳，第一群最差。

La première partie de cette thèse est consacrée aux mélanges convolutifs (sur-)déterminés. Nous cherchons à étendre l'algorithme FastICA aux mélanges convolutifs en proposant un algorithme à point fixe opérant dans le domaine temporel. Nous introduisons un processus de blanchiment spatio-temporel non-causal, qui, en initialisant les paramètres d'extraction d'une façon particulière, permet d'utiliser des itérations d'optimisation de type point fixe. L'estimation des contributions dans les observations est réalisée grâce à un critère quadratique optimisé par un filtre de Wiener non-causal.<br /> Dans la deuxième partie, consacrée aux mélanges sous-déterminés instantanés et convolutifs, nous cherchons aussi à étendre l'algorithme FastICA en nous basant sur le concept de séparation différentielle. Nous proposons des procédures de blanchiment différentiel des observations qui permettent l'emploi d'itérations à point fixe pour séparer entre elles des sources dites utiles. En convolutif, les contributions sont estimées au moyen d'un filtre de Wiener différentiel non-causal. <br /> La troisième partie, consacrée aux mélanges instantanés de sources contenant un grand nombre d'échantillons, met d'abord en évidence une limitation de l'algorithme FastICA, qui, à chaque itération d'optimisation, calcule des statistiques sur l'ensemble des échantillons disponibles. Le critère du kurtosis étant polynômial relativement aux coefficients d'extraction, nous proposons une procédure d'identification de ce polynôme qui permet d'effectuer l'optimisation dans un espace de calcul moins gourmand. Ce principe est ensuite appliqué aux mélanges instantanés sous-déterminés.

The purpose of this study was to examine the factorial validity and measurement equivalence of the Test of Performance Strategies (TOPS; Thomas, Murphy, & Hardy, 1999); the Sport Anxiety Scale (SAS; Smith, Smoll, & Schultz, 1990); and the Golf Performance Survey (GPS; Thomas & Over, 1994) across age groups in a representative sample of amateur golfers. Based on archival data, participants comprising this study were 649 younger adult (n = 237) and older adult (n = 412) amateur golfers who played in the Dupont World Amateur Golf Championship in Myrtle Beach, South Carolina. The participants completed a set of questionnaires including psychological skills and strategies (e.g., self-talk, goal setting, imagery, etc.) used during competition, sport-specific competitive trait anxiety, and psychomotor skills and involvement in golf. Results demonstrated that the original factor structure of the TOPS competition subscale, the SAS, and the GPS, did not adequately fit the data among this sample of younger and older adult amateur golfers. Further exploratory and confirmatory factor analyses established evidence of factorial validity with the TOPS competition subscale, SAS, and the GPS with both younger and older adult amateur golfers. Configural, metric, scalar, and strict measurement invariance were identified in relation to the TOPS competition subscale, SAS, and the GPS across age cross-group comparisons. In general, the analyses demonstrated support that the TOPS competition subscale, SAS, and the GPS can be utilized with confidence with older adult amateur golfers, as well as conducting group comparisons with younger adult amateur golfers. The findings from this study have several future research directions and practical implications for structuring effective interventions with older adult amateur athletes.

L'objectif de ce travail était de tester des méthodes de SAS pour la séparation des spectres complexes RMN de mélanges dans les plus simples des composés purs. Dans une première partie, les méthodes à savoir JADE et NNSC ont été appliqué es dans le cadre de la DOSY , une application aux données CPMG était démontrée. Dans une deuxième partie, on s'est concentré sur le développement d'un algorithme efficace "beta-SNMF" . Ceci s'est montré plus performant que NNSC pour beta inférieure ou égale à 2. Etant donné que dans la littérature, le choix de beta a été adapté aux hypothèses statistiques sur le bruit additif, une étude statistique du bruit RMN de la DOSY a été faite pour obtenir une image plus complète de nos données RMN étudiées. / The objective of the work was to test BSS methods for the separation of the complex NMR spectra of mixtures into the simpler ones of the pure compounds. In a first part, known methods namely JADE and NNSC were applied in conjunction for DOSY , performing applications for CPMG were demonstrated. In a second part, we focused on developing an effective algorithm "beta- SNMF ". This was demonstrated to outperform NNSC for beta less or equal to 2. Since in the literature, the choice of beta has been adapted to the statistical assumptions on the additive noise, a statistical study of NMR DOSY noise was done to get a more complete picture about our studied NMR data.

Experimentos cujas variaveis respostas s~ ao proporcoes ou contagens, sao muito comuns nas diversas areas do conhecimento, principalmente na area agricola. Na analise desses experimentos, utiliza-se a teoria de modelos lineares generalizados, bastante difundida (McCullagh & Nelder, 1989; Demetrio, 2001), em que as respostas sao independentes. Caso a variancia estimada seja maior do que a esperada, estima-se o parametro de dispersao, incluindo-o no processo de estimaçao dos parametros. Quando a variavel resposta e observada ao longo do tempo, pode haver uma correlacao entre as observacoes e isso tem que ser levado em consideracao na estimacao dos parametros. Uma forma de se trabalhar essa correlacao e aplicando a metodologia de equacoes de estimacao generalizada (EEG), discutida por Liang & Zeger (1986), embora, neste caso, o interesse esteja nas estimativas dos efeitos fixos e a inclusao da matriz de correlacao de trabalho sirva para se obter um melhor ajuste. Uma outra alternativa e a inclusao, no preditor linear, de um efeito latente para captar variabilidades nao consideradas no modelo e que podem in uenciar nos resultados. No presente trabalho, usa-se uma forma combinada de efeito aleatorio e parametro de dispersao, incluidos conjuntamente na estimacao dos parametros. Essa metodologia e aplicada a um conjunto de dados obtidos de um experimento com camu-camu, com objetivo de se avaliarem quais os melhores metodos de enxertia e tipos de porta-enxertos que podem ser utilizados, atraves da proporcao de pegamentos da muda. Varios modelos sao ajustados, desde o modelo em parcelas subdivididas (supondo independencia), ate o modelo em que se considera o parametro de dispersao e efeito aleatorio conjuntamente. Ha evidencias de que o modelo em que se inclui o efeito aleatorio e o parametro de dispersao, conjuntamente, resultam em melhores estimativas dos parametros. Outro conjunto de dados longitudinais, com milho transgenico MON810, em que a variavel resposta e o numero de lagartas (Spodoptera frugiperda), e utilizado. Neste caso, devido ao excesso de respostas zero, emprega-se o modelo de regressao Poisson in acionado de zeros (ZIP), alem do modelo Poisson padrao, em que as observacoes sao consideradas independentes, e do modelo Poisson in acionado de zeros com efeito aleatorio. Os resultados mostram que o efeito aleatorio incluido no preditor foi nao significativo e, assim, o modelo adotado e o modelo de regressao Poisson in acionado de zeros. Os resultados foram obtidos usando-se os procedimentos NLMIXED, GENMOD e GPLOT do SAS - Statistical Analysis System, versao 8.2. / Experiments which response variables are proportions or counts are very common in several research areas, specially in the area of agriculture. The theory of generalized linear models, well difused (McCullagh & Nelder, 1989; Demetrio, 2001), is used for analyzing these experiments where the responses are independent. If the estimated variance is greater than the expected variance, the dispersion parameter is estimated including it on the parameter estimation process. When the response variable is observed over time a correlation among observations might occur and it should be taken into account in the parameter estimation. A way of dealing with this correlation is applying the methodology of generalized estimating equations (GEEs) discussed by Liang & Zeger (1986) although, in this case, the interest is on the estimates of the xed efect being the inclusion of a working correlation matrix useful to obtain more accurate estimates. Another alternative is the inclusion of a latent efect in the linear predictor to explain variabilities not considered in the model that might in uence the results. In this work the random efect and the dispersion parameter are combined and included together in the parameter estimation. Such methodology is applied to a data set obtained from an experiment realized with camu-camu to evaluate, through proportion of grafting well successful of seedling, which kind of grafting and understock are suitable to be used. Several models are fitted, since the split plot model (with independence assumption) up to the model where the dispersion parameter and the random efect are considered together. There is evidence that the model including the random efect and the dispersion parameter together, produce better estimates of the parameters. Another longitudinal data set used here comes from an experiment realized with the MON810 transgenic corn where the response variable is the number of caterpillars (Spodoptera frugiperda). In this case, due to the excessive number of zeros obtained, the zero in ated Poisson regression model (ZIP) is used in addition to the standard Poisson model, where observations are considered independent, and the zero in ated Poisson regression model with random efect. The results show that the random efect included in the linear predictor was not significant and, therefore, the adopted model is the zero in ated Poisson regression model. The results were obtained using the procedures NLMIXED, GENMOD and GPLOT available on SAS - Statistical Analysis System, version 8.2.