The ROC (Receiver Operating Characteristic) curve is a projection of two different cumulative distribution functions F0 and F1. On axis are values 1-F0(c) and 1-F1(c). The c-parameter is a real number. This curve is useful to check quality of discriminant rule which classify an object to one of two classes. The criterion is a size of an area under the curve. To solve real problems we use point and interval estimation of ROC curves and statistical hypothesis tests about ROC curves.

Orientador: Francisco José Teixeira Neto / Resumo: Objetivo: Comparar a acurácia diagnóstica da variação da pressão de pulso (VPP), variação do volume sistólico mensurada através do análise de contorno de pulso (VVSACP), índice de variabilidade pletismográfica (IVP), pressão venosa central (PVC) e índice do volume diastólico final global mensurado pela técnica de termodiluição transpulmonar (GEDVITDTP) para predizer a fluido-responsividade em cães. Animais: Quarenta cadelas saudáveis (13,8–26,8 kg) submetidas a ovário-salpingo-histerectomia. Métodos: A anestesia foi mantida com isoflurano sob ventilação mecânica com volume controlado (volume corrente 12 mL/kg, pausa inspiratória 40%, relação inspiração/expiração: 1:1,5). O débito cardíaco e o volume sistólico foram obtidos pela técnica de termodiluição transpulmonar através de um cateter inserido na artéria femoral. A fluido-responsividade (FR) foi avaliada por uma prova de carga (solução de Ringer com lactato, 20 mL/kg durante 15 minutos), administrada uma vez (n = 21) ou duas vezes (n = 18) antes da cirurgia. Respondedores a volume foram definidos como indivíduos onde o VS mensurado pela técnica de termodiluição transpulmonar, elevou-se acima de 15% após a última prova de carga. Resultados: Dos 39 animais incluídos no estudo, 21 cães foram classificados como respondedores e 18 não respondedores ao último desafio volêmico. As áreas sob as curvas de características de operação do receptor (AUROC) foram de 0,976, 0,906, 0,868 e 0,821 para VPP, IVP, PVC e VVSACP, respectivament... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Objective: To compare the diagnostic accuracy of pulse pressure variation (PPV), stroke volume variation from pulse contour analysis (SVVPCA), plethysmographic variability index (PVI), central venous pressure (CVP) and global end-diastolic volume index measured by transpulmonary thermodilution (GEDVITPTD) to predict fluid responsiveness (FR) in dogs. Animals: A group of 40 bitches (13.8–26.8 kg) undergoing ovariohysterectomy. Methods: Anesthesia was maintained with isoflurane under volume-controlled ventilation (tidal volume 12 mL/kg; inspiratory pause during 40% of inspiratory time; inspiration: expiration ratio 1:1.5). Transpulmonary thermodilution cardiac output was recorded through a femoral artery catheter. FR was evaluated by a fluid challenge (lactated Ringer's, 20 mL kg over 15 minutes) administered once (n = 21) or twice (n = 18) before surgery. Individuals were responders if stroke volume index measured by transpulmonary thermodilution increased >15% after the last fluid challenge. Results: Of the 39 animals studied, 21 were responders and 18 were nonresponders. Area under the receiver operating characteristics curve (AUROC) was 0.976, 0.906, 0.868 and 0.821 for PPV, PVI, CVP and SVVPCA, respectively (p < 0.0001 from AUROC = 0.5). GEDVITPTD failed to predict FR (AUROC: 0.660, p = 0.078). Best cut-off thresholds discriminating responders and nonresponders, with respective zones of diagnostic uncertainty (gray zones) were: PPV >16% (15–16%), PVI >11% (10–13%), SVVPCA ... (Complete abstract click electronic access below) / Doutor

Background: To compare the clinical value of the measurement of complex and total PSA in the discrimination between benign prostatic hyperplasia (BPH) and prostate cancer. Methods: In serum samples collected from 166 men with histopathologically proven clinically localized prostate cancer and of 97 men with BPH, total prostate-specific antigen (PSA), complexed PSA and the free to total PSA ratio were determined. The statistical analysis was done by the comparison of the receiver operator characteristic (ROC) curves. Results: The areas under the ROC curves were 0.776 for total PSA, 0.799 for complexed PSA (total PSA vs. cPSA: p < 0.0001) and 0.812 for the free to total PSA ratio. With a cut-off of 3.0 ng/ml for complexed PSA, the sensitivity was 90%, the specificity 58%, the positive and the negative predictive values 79 and 78%, respectively. With a cut-off of 4.0 ng/ml for total PSA, the sensitivity was 87%, the specificity 59%, the positive and the negative predictive values were 78 and 72%, respectively. Conclusions: There was a statistically significant advantage for complexed PSA compared to total PSA in the discrimination between BPH and prostate cancer. The difference was, however, small and its clinical relevance is questionable. / Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG-geförderten) Allianz- bzw. Nationallizenz frei zugänglich.

Der Großteil psychischer Störungen wird derzeit auf Grundlage von an Symptomen orientierten Verhaltensbeschreibungen diagnostiziert. Objektive Biomarker könnten die Diagnostik und Behandlung psychischer Störungen verbessern. Der Zusammenhang zwischen Geruchssinn und Depressionen wurde erstmals bei Nagetieren beschrieben und bald darauf auch beim Menschen. Daraufhin wurde diskutiert, ob der Geruchssinn als Biomarker der Erkrankung dienen könnte. Neben einer reduzierten Riechfunktion zeigen Betroffene von Depressionen ebenfalls ein reduziertes Volumen des Bulbus Olfactorius (BO). Bisher war jedoch unklar, in wie fern die Reduktion spezifisch für Depressionen ist und damit als diagnostischer Marker genutzt werden könnte. Weiterhin offen war, ob diese Reduktion des Volumens des BO von Veränderungen in anderen Hirnstrukturen moderiert wird. Mit dem Ziel der Klärung dieser Fragen, wurden 84 PatientInnen der Klinik und Poliklinik für Psychotherapie und Psychosomatik des Carl Gustav Carus Universitätsklinikums in Dresden und 51 altersangepasste psychisch gesunde KontrollprobandInnen rekrutiert. Es erfolgten Untersuchungen im Magnetresonanztomographen und das individuelle Volumen des BO wurde aus diesen Daten manuell berechnet und zwischen den Gruppen verglichen. Außerdem wurden eine multiple Regressionsanalyse mit dem BO Volumen als abhängige Variable und eine Receiver Operater Characteristic Analyse, mit dem Ziel die diagnostische Genauigkeit des BO für Depressionen zu erfassen, durchgeführt. Eine voxelbasierte morphometrische Analyse des gesamten Gehirns offenbarte Hirnstrukturen, die sowohl mit Depressionen als auch dem Volumen des BO zusammenhingen. Im Ergebnis zeigte die Gruppe der Patienten ein durchschnittlich um 13.5% reduziertes Volumen des BO (d=.52). Die Regressionsanalyse zeigte, dass das Volumen des BO am besten von der Kombination der Variablen Depression (β=-.19), Geschlecht (β=-.31) und Alter (β=-.29) vorhergesagt wurde. Keine andere psychische Störung erreichte in diesem Regressionsmodell Signifikanz. Die diagnostische Genauigkeit des BO Volumens für die Diagnose einer Depression erreichte 68.1%. Das Volumen des BO war assoziiert mit dem Volumen der grauen Substanz in folgenden Arealen: insulärer Kortex (bilateral), rechte Amygdala, Hippokampus (bilateral), Gyrus Rectus (bilateral), superiorer temporaler Gyrus (STG, bilateral) und piriformer Gyrus (bilateral). Hirnstrukturen mit reduziertem Volumen der grauen Substanz bei Vorliegen einer Depression waren das rechte Kleinhirn, rechter Thalamus, linker angularer Gyrus, rechter okzipitaler inferiorer Gyrus und ein Cluster, dass den linken insulären Kortex und den linken STG umfasste. Der Zusammenhang zwischen 4 dem Volumen des BO und Depressionen wurde von Volumenreduktionen der grauen Substanz im letztgenannten Cluster moderiert. Zusammenfassend zeigten die Untersuchungen eine Volumenreduktion des BO von mittlerer Effektstärke bei Patienten mit psychischen Störungen und von allen psychischen Störungen war diese am stärksten mit der Diagnose einer Depression verknüpft. Folglich kann der BO als ein Marker für Depressionen gelten und komplettiert damit jene Hirnstrukturen, die typischerweise als bei Depressionen verändert diskutiert werden, wie beispielsweise der Hippokampus, die Amygdala, der anteriore cinguläre Kortex und präfrontale Areale. Allerdings zeigte der BO als alleiniges Kriterium für die Diagnose einer Depression keine befriedigende diagnostische Genauigkeit zur Nutzung im klinischen Alltag. Hier könnte die Kombination mit weiteren strukturellen Hirnmarkern die Genauigkeit möglicherweise erhöhen. Das Cluster, das den Zusammenhang zwischen dem Volumen des BO und Depressionen moderierte beinhaltete zwei Areale, die dem Salienznetzwerk zugeschrieben werden. Dies unterstützt Theorien, welche Salienzerkennung, (olfaktorische) Aufmerksamkeit und Depressionen in Zusammenhang bringen. Eine Klärung der kausalen Mechanismen dieser Zusammenhänge ist aufgrund des Querschnittsdesigns der Untersuchung jedoch nicht möglich. Deswegen werden in der Arbeit zwar mögliche Mechanismen diskutiert, zur Klärung sind aber längsschnittliche Untersuchungen notwendig.:Zusammenfassung 3 Summary 5 Hintergrund 7 Studienziele 10 Material und Methoden 11 Ergebnisse 13 Schlussfolgerungen 15 Literaturverzeichnis 18 Veröffentlichungen, Konferenzbeiträge und andere Leistungen 22 Anhang 24 Erklärungen zur Eröffnung des Promotionsverfahrens 26 Danksagung 29

Au Québec, il existe près de 6000 barrages et ouvrages régulateurs dont il est nécessaire d'assurer la sécurité et la stabilité. Les barrages-poids en béton, concernés par cette étude, sont soumis à plusieurs sollicitations telles que la poussée des eaux et l'action des sous-pressions. Ces sollicitations peuvent entraîner le cisaillement de l'ouvrage selon qu'elles se développent au sein d'une ou plusieurs interfaces suivantes: contact béton-béton, interface béton-roc de fondation et discontinuités roc-roc. La résistance au cisaillement de ces matériaux est gouvernée par leur friction et leur cohésion. La détermination de ces paramètres est donc primordiale dans l'établissement de la stabilité d'un barrage-poids en béton. Pour répondre à cette problématique, ce projet de maîtrise s'est concentré sur les paramètres de résistance au cisaillement (principalement l'angle de friction et la cohésion) des différentes interfaces des barrages, obtenus à partir des essais de résistance au cisaillement. Une base de données comportant plusieurs centaines d'essais a été constituée à partir de rapports d'essais de résistance au cisaillement disponibles sur des barrages-poids en béton du Québec.Au total, des essais provenant de 30 barrages-poids en béton, 4 carrières et 2 projets de barrages ont été incorporés dans la base de données. Ces ouvrages, construits entre 1911 à aujourd'hui, ont été investigués dans les années 1985 à 2010. Les données recueillies ont permis d'établir que plusieurs variables semblaient influencés [i.e. influencées] par la résistance au cisaillement. Les facteurs influençant les paramètres de résistance au cisaillement (tels que la nature du roc, la rugosité, l'effet d'échelle, la granulométrie, l'altération, etc.) ont été étudiés par des analyses statistiques. Notamment, les analyses en composantes principales ont permis d'établir des corrélations entre les différentes variables et les données d'essais, tandis que les classifications ascendantes hiérarchiques ont regroupé les données semblables et contribuer [i.e. contribué] à mieux comprendre les relations entre les variables. Plus particulièrement, ce sont les relations entre les paramètres de résistance au cisaillement et les types de roches formant les fondations rocheuses des barrages du Québec qui étaient recherchés [i.e. recherchées]. Les principaux résultats de ces analyses ont été les suivants : (1) Il n'existe pas de relation linéaire (c'est-à-dire directe) entre les paramètres de résistance au cisaillement et les divers types de rocher de fondation ; (2) Il n'y a pas d'effet d'échelle sur les résultats à l'échelle de l'éprouvette de laboratoire ; (3) Il existe une corrélation importante entre les paramètres de résistance au cisaillement et les paramètres de rugosité des épontes. Cette étude a donc permis de mieux comprendre le comportement au cisaillement des interfaces d'un barrage-poids en béton et d'établir que le facteur prépondérant demeure relié à la morphologie des épontes des discontinuités. En outre, certaines modifications à la méthodologie actuelle pour la détermination des paramètres de résistance au cisaillement ont été proposées, permettant d'approfondir les relations avec la rugosité des épontes.

In medical diagnostic studies, the area under the Receiver Operating Characteristic (ROC) curve (AUC) and Youden index are two summary measures widely used in the evaluation of the diagnostic accuracy of a medical test with continuous test results. The first half of this dissertation will highlight ROC analysis including extension of Youden index to the partial Youden index as well as novel confidence interval estimation for AUC and Youden index in the presence of covariates in induced linear regression models. Extensive simulation results show that the proposed methods perform well with small to moderate sized samples. In addition, some real examples will be presented to illustrate the methods. The latter half focuses on the application of empirical likelihood method in economics and finance. Two models draw our attention. The first one is the predictive regression model with independent and identically distributed errors. Some uniform tests have been proposed in the literature without distinguishing whether the predicting variable is stationary or nearly integrated. Here, we extend the empirical likelihood methods in Zhu, Cai and Peng (2014) with independent errors to the case of an AR error process. The proposed new tests do not need to know whether the predicting variable is stationary or nearly integrated, and whether it has a finite variance or an infinite variance. Another model we considered is a GARCH(1,1) sequence or an AR(1) model with ARCH(1) errors. It is known that the observations have a heavy tail and the tail index is determined by an estimating equation. Therefore, one can estimate the tail index by solving the estimating equation with unknown parameters replaced by Quasi Maximum Likelihood Estimation (QMLE), and profile empirical likelihood method can be employed to effectively construct a confidence interval for the tail index. However, this requires that the errors of such a model have at least finite fourth moment to ensure asymptotic normality with n1/2 rate of convergence and Wilk's Theorem. We show that the finite fourth moment can be relaxed by employing some Least Absolute Deviations Estimate (LADE) instead of QMLE for the unknown parameters by noting that the estimating equation for determining the tail index is invariant to a scale transformation of the underlying model. Furthermore, the proposed tail index estimators have a normal limit with n1/2 rate of convergence under minimal moment condition, which may have an infinite fourth moment, and Wilk's theorem holds for the proposed profile empirical likelihood methods. Hence a confidence interval for the tail index can be obtained without estimating any additional quantities such as asymptotic variance.

The Receiver Operating Characteristic (ROC) curve is the plot of Sensitivity vs. 1- Specificity of a quantitative diagnostic test, for a wide range of cut-off points c. The empirical ROC curve is probably the most used nonparametric estimator of the ROC curve. The asymptotic properties of this estimator were first developed by Hsieh and Turnbull (1996) based on strong approximations for quantile processes. Jensen et al. (2000) provided a general method to obtain regional confidence bands for the empirical ROC curve, based on its asymptotic distribution. Since most biomarkers do not have high enough sensitivity and specificity to qualify for good diagnostic test, a combination of biomarkers may result in a better diagnostic test than each one taken alone. Su and Liu (1993) proved that, if the panel of biomarkers is multivariate normally distributed for both diseased and non-diseased populations, then the linear combination, using Fisher's linear discriminant coefficients, maximizes the area under the ROC curve of the newly formed diagnostic test, called the generalized ROC curve. In this dissertation, we will derive the asymptotic properties of the generalized empirical ROC curve, the nonparametric estimator of the generalized ROC curve, by using the empirical processes theory as in van der Vaart (1998). The pivotal result used in finding the asymptotic behavior of the proposed nonparametric is the result on random functions which incorporate estimators as developed by van der Vaart (1998). By using this powerful lemma we will be able to decompose an equivalent process into a sum of two other processes, usually called the brownian bridge and the drift term, via Donsker classes of functions. Using a uniform convergence rate result given by Pollard (1984), we derive the limiting process of the drift term. Due to the independence of the random samples, the asymptotic distribution of the generalized empirical ROC process will be the sum of the asymptotic distributions of the decomposed processes. For completeness, we will first re-derive the asymptotic properties of the empirical ROC curve in the univariate case, using the same technique described before. The methodology is used to combine biomarkers in order to discriminate lung cancer patients from normals.

最近大韓民國的報章雜誌上常有關於中華民國各方面快速進步的報導，尤其在經濟建設方面，大規模工廠的創設，對外貿易的推進都有令人驚訝的長足進展。 中華民國是韓國的近鄰，也是我們反共的堅強盟友，對於中華民國的發展，自為我國人士所樂聞，而中華民國工商貿易的發展，其實際情形如何，對於我國經濟發展會有什麼影響，也為我國政府人士所關切。 然而在收集資料的過程中，因為歷年來兩國的貿易統計數目有很大的差異，所以存在著許多困難的問題，如有不夠詳細與完美的地方尚請各位老師多多指教。 此論文顧名思義以中韓貿易關係的研究為主題，並加以詳細分析與討論。 由於科學昌明，促成了經濟的迅速發展，由於交通發達，促進了國際間交往的頻繁。再加上三種生產費差異的作用，促成了經濟上的國際分工，中韓兩國的貿易也因而更見繁盛，為了有助於積極並順利地發展中韓國際貿易與共同從事於國際市場的開發，促成我在企業管理研究所專攻行銷學研究及作此論文的動機。 其目的在將一九四五年至一九七O年間中韓經濟發展及對外貿易的情形加以描述，以供中華民國政府人士及學術界做為參考資料，如果能在增進中韓貿易和貿易均衡上有萬分之一的助益，將是我最高的榮幸。 中韓貿易額近生來呈現增加的趨勢，就中華民國對大韓民國的輸出而言，雖各年互有增減，但長期觀之，則有增加的趨勢。至去年（一九七O年）已經至三千萬美元以上，就對大韓民國之輸入而言，仍呈現增加的趨勢，但如果我們從輸出輸入所佔比率上觀之，則中華民國對大韓民國的輸出，佔輸出總值的比率顯示不穩定的變動，也就是說，對韓國的輸出並沒有比例增加，並且所佔比值很小，約僅百分之二。中華民國對大韓民國之輸入比值更小，約僅佔輸入總值的千分之五。 如從貿易額所佔之席位觀之，則中華民國對大韓民國之輸出至一九七０生佔對大韓民國輸出的第九位，而大韓民國對中華民國之輸出佔第十位。 從上述分析可知；中韓雙方雖努力於貿易的促進，然而其成果卻不大。其原因乃是中韓工商業發展的方向極為相似，並且市場狹小，國內市場極易飽和，怎能夠容許國外廠商進入。故近年來，主要貿易項目乃局限於自然氣候影饗較大的農產品。 結論：(1)市場共享：以兩國市場為對象，設立較大而合乎經濟規模之各項工業，促使兩國經建順利發展，且可互增輸出。 (2)截長補短：兩國政府成業者，並該預先做市場調查，使過多或不足之產品以國際合理的價格相互購售，避免貨物滯銷造成浪費現象。 (3)捷足先登：視中韓市場需要而生產最必要的物品，以物美價廉與他國競爭。 (4)互易有無：中韓兩國地理因素完全不同，生產品也隨之有異，應該藉國際貿易互通有無。 (5)先洽後展：雙方經營貿易者應該先彼此獲得了解與信任，經由有關單位發展貿易，裨使解除彼此間對日貨的依賴。

Approved for public release, distribution is unlimited / Since 1988, democracy in Taiwan has evolved and developed a great deal. Experts argue whether this growth constitutes "democratic consolidation" but there is no contention of the idea that the ROC is more democratic now versus pre-1988. In addition, public opinion polls show that the populace view themselves very differently in 2004 than they did in 1988 in terms of their national identity and their preferences for mainland relations. Finally, the democratic period in Taiwan witnessed greater hostility between the PRC and ROC than in the preceding thirty years combined. This heightened level of belligerence has subsided in recent years, but still remains a Sword of Damocles hanging over each step of Taiwan's democratic process. With these facts in mind, it is clear that the addition of ROC democratization has destabilized relations between the China and Taiwan. One should note that Taiwan's political liberalization has not damned cross-strait relations to a cataclysmic fate. As noted many times in this essay there are prospects for hope and increased cooperation. However, with the advent of democracy for the first time in an ethnically Chinese society, relations moved from a fairly stable equilibrium to a somewhat chaotic new reality resplendent with uncertainty and ripe for catastrophic miscalculation. / Captain, United States Air Force

Feature selection method is becoming an essential procedure in data preprocessing step. The feature selection problem can affect the efficiency and accuracy of classification models. Therefore, it also relates to whether a classification model can have a reliable performance. In this study, we compared an original feature selection method and a proposed frequency-based feature selection method with four classification models and three filter-based ranking techniques using a cancer dataset. The proposed method was implemented in WEKA which is an open source software. The performance is evaluated by two evaluation methods: Recall and Receiver Operating Characteristic (ROC). Finally, we found the frequency-based feature selection method performed better than the original ranking method.