No description available.

No description available.

銀行在評分模型建置完成後需進行驗證工作，以瞭解評分模型是否能有效評出客戶的風險層級，穩健地估計區別鑑別力指標為驗證工作中的重點。在先前的文獻中假設正常授信戶與違約戶分數分配為常態分配。但在實際資料中，分配未必定為常態。因此本文接著探討在正常授信戶與違約授信戶之分配為混合分配，即兩分數分配為偏斜常態分配下，何種方法可以對於估計AUC具有較高的穩定性。本文比較五種估計AUC的方法，分別為常態核，經驗分配，曼惠尼近似，最大摡似法和EM演算法。模擬結果呈現(1)投信戶組合分配為兩常態分配下，最大摡似法在大部分違約率下都可以得到較窄的信賴區間。(2)組合分配為一常態與一偏斜常態及兩偏斜常態分配下，EM演算法在大部分情況有較窄的信賴區間，其中在兩偏斜常態分配下，表現更佳。(3)曼惠尼近似建構的信賴區間寬度最大，代表曼惠尼近似是較保守的估計方法。 / Banks face discrimination after constructing the rating systems to figure out whether the systems can discriminate defaulting and non-defaulting borrowers. Literature assumed the two score distribuion are normal distributed. However, the real data may not be normal distribuions. We assum the two score distribuions are skewed normal distribuions to discuss which method has more robustness to estimate the AUC value.Under skewed distribution, we propose EM algorithm to estimate the population parametric. If used properly, information about the population properties may be used to get better accuracy of estimation the AUC value.Numerical results show the EM algorithm method , comparing with other methods, has robustness in detect the rating systems have discirmatory power.

This thesis aims to discern whether cyclical history can be appropriately applied to the Ab Urbe Condita, and from these findings discern Livy's authorial implications for the contemporary political program. This process is conducted by analyzing exempla , as well as constructing a new definition of metus hostilis. Doing so allows for the detection of patterns, that when imprinted upon the existing formulaic model, examines whether the metus hostilis enhances the case for Livy writing the AUC with cyclical intent. Based on this analysis, the implications for contemporary Rome are clear in that the narrative insinuates the Augustan regime's necessity.

Doctor of Philosophy / Department of Biochemistry / Ramaswamy Krishnamoorthi / Insect [beta]-1,3-glucan recognition protein ([beta]GRP), a soluble receptor in the hemolymph, binds to the surfaces of bacteria and fungi and activates serine protease cascades that promote destruction of pathogens by means of melanization or expression of antimicrobial peptides. Delineation of mechanistic details of these processes may help develop strategies to control insect-borne diseases and economic losses. Multi-dimensional nuclear magnetic resonance (NMR) techniques were employed to solve the solution structure of the Indian meal moth (Plodia interpunctella) [beta]GRP N-terminal domain (N-[beta]GRP), which is sufficient to activate the prophenoloxidase (proPO) pathway resulting in melanin formation. This is the first determined three-dimensional structure of N-[beta]GRP, which adopts an immunoglobulin fold. Addition of laminarin, a [beta]-1,3 and [beta]-1,6 link-containing glucose polysaccharide (∼6 kDa) that activates the proPO pathway, to N-[beta]GRP results in the loss of NMR cross-peaks from the backbone [subscript]1[subscript]5N-[subscript]1H groups of the protein, suggesting the formation of a large complex. Analytical ultracentrifugation (AUC) studies of formation of the N-[beta]GRP:laminarin complex show that ligand binding induces self-association of the protein-carbohydrate complex into a macro structure, likely containing six protein and three laminarin molecules (∼102 kDa). The macro complex is quite stable, as it does not undergo dissociation upon dilution to submicromolar concentrations. The structural model thus derived from this study for the N-[beta]GRP:laminarin complex in solution differs from the one in which a single N-[beta]GRP molecule has been proposed to bind to a triple-helical form of laminarin on the basis of a X-ray crystal structure of the N-[beta]GRP:laminarihexaose complex. AUC studies and phenoloxidase activation measurements made with designed mutants of N-[beta]GRP indicate that electrostatic interactions between the ligand-bound protein molecules contribute to the stability of the N-[beta]GRP:laminarin complex and that a decreased stability results in a reduction of proPO activation. These novel findings suggest that ligand-induced self-association of the [beta]GRP:[beta]-1,3-glucan complex may form a platform on a microbial surface for recruitment of downstream proteases, as a means of amplification of the pathogen recognition signal. In the case of the homolog of GNBPA2 from Anopheles gambiae, the malaria-causing Plasmodium carrier, multiligand specificity was characterized, suggesting a functional diversity of the immunoglobulin domain structure.

The receiver operating characteristic (ROC) curve methodology is the statistical methodology for assessment of the accuracy of diagnostics tests or bio-markers. Currently most widely used statistical methods for the inferences of ROC curves are complete-data based parametric, semi-parametric or nonparametric methods. However, these methods cannot be used in diagnostic applications with missing data. In practical situations, missing diagnostic data occur more commonly due to various reasons such as medical tests being too expensive, too time consuming or too invasive. This dissertation aims to develop new nonparametric statistical methods for evaluating the accuracy of diagnostic tests or biomarkers in the presence of missing data. Specifically, novel nonparametric statistical methods will be developed with different types of missing data for (i) the inference of the area under the ROC curve (AUC, which is a summary index for the diagnostic accuracy of the test) and (ii) the joint inference of the sensitivity and the specificity of a continuous-scale diagnostic test. In this dissertation, we will provide a general framework that combines the empirical likelihood and general estimation equations with nuisance parameters for the joint inferences of sensitivity and specificity with missing diagnostic data. The proposed methods will have sound theoretical properties. The theoretical development is challenging because the proposed profile log-empirical likelihood ratio statistics are not the standard sum of independent random variables. The new methods have the power of likelihood based approaches and jackknife method in ROC studies. Therefore, they are expected to be more robust, more accurate and less computationally intensive than existing methods in the evaluation of competing diagnostic tests.

To significantly improve the early detection of cervical precancers and cancers, LightTouch™ is under development by SpectRx Inc.. LightTouch™ identifies cancers and precancers quickly by using a spectrometer to analyze light reflected from the cervix. Data from the spectrometer is then used to create an image of the cervix that highlights the location and severity of disease. Our research is conducted to find the appropriate models that can be used to generate map-like image showing disease tissue from normal and further diagnose the cervical cancerous conditions. Through large work of explanatory variable search and reduction, logistic regression and Partial Least Square Regression successfully applied to our modeling process. These models were validated by 60/40 cross validation and 10 folder cross validation. Further examination of model performance, such as AUC, sensitivity and specificity, threshold had been conducted.

Receiver Operating Characteristic (ROC) curves are often used to evaluate the prognostic performance of a continuous biomarker. In a previous research, a non-parametric ROC approach was introduced to compare two biomarkers with repeated measurements. An asymptotically normal statistic, which contains the subject-specific weights, was developed to estimate the areas under the ROC curve of biomarkers. Although two weighting schemes were suggested to be optimal when the within subject correlation is 1 or 0 by the previous study, the universal optimal weight was not determined. We modify this asymptotical statistic to compare AUCs between two correlated groups and propose a solution to weight optimization in non-parametric AUCs comparison to improve the efficiency of the estimator. It is demonstrated how the Lagrange multiplier can be used as a strategy for finding the weights which minimize the variance function subject to constraints. We show substantial gains of efficiency by using the novel weighting scheme when the correlation within group is high, the correlation between groups is high, and/or the disease incidence is small, which is the case for many longitudinal matched case-control studies. An illustrative example is presented to apply the proposed methodology to a thyroid function dataset. Simulation results suggest that the optimal weight performs well with a sample size as small as 50 per group.

The Top Scoring Pair (TSP) classifier, based on the notion of relative ranking reversals in the expressions of two marker genes, has been proposed as a simple, accurate, and easily interpretable decision rule for classification and class prediction of gene expression profiles. We introduce the AUC-based TSP classifier, which is based on the Area Under the ROC (Receiver Operating Characteristic) Curve. The AUCTSP classifier works according to the same principle as TSP but differs from the latter in that the probabilities that determine the top scoring pair are computed based on the relative rankings of the two marker genes across all subjects as opposed to for each individual subject. Although the classification is still done on an individual subject basis, the generalization that the AUC-based probabilities provide during training yield an overall better and more stable classifier. Through extensive simulation results and case studies involving classification in ovarian, leukemia, colon, and breast and prostate cancers and diffuse large b-cell lymphoma, we show the superiority of the proposed approach in terms of improving classification accuracy, avoiding overfitting and being less prone to selecting non-informative pivot genes. The proposed AUCTSP is a simple yet reliable and robust rank-based classifier for gene expression classification. While the AUCTSP works by the same principle as TSP, its ability to determine the top scoring gene pair based on the relative rankings of two marker genes across {\em all} subjects as opposed to each individual subject results in significant performance gains in classification accuracy. In addition, the proposed method tends to avoid selection of non-informative (pivot) genes as members of the top-scoring pair.\\ We have also proposed the use of the AUC test statistic in order to reduce the computational cost of the TSP in selecting the most informative pair of genes for diagnosing a specific disease. We have proven the efficacy of our proposed method through case studies in ovarian, colon, leukemia, breast and prostate cancers and diffuse large b-cell lymphoma in selecting informative genes. We have compared the selected pairs, computational cost and running time and classification performance of a subset of differentially expressed genes selected based on the AUC probability with the original TSP in the aforementioned datasets. The reduce sized TSP has proven to dramatically reduce the computational cost and time complexity of selecting the top scoring pair of genes in comparison to the original TSP in all of the case studies without degrading the performance of the classifier. Using the AUC probability, we were able to reduce the computational cost and CPU running time of the TSP by 79\% and 84\% respectively on average in the tested case studies. In addition, the use of the AUC probability prior to applying the TSP tends to avoid the selection of genes that are not expressed (``pivot'' genes) due to the imposed condition. We have demonstrated through LOOCV and 5-fold cross validation that the reduce sized TSP and TSP have shown to perform approximately the same in terms of classification accuracy for smaller threshold values. In conclusion, we suggest the use of the AUC test statistic in reducing the size of the dataset for the extensions of the TSP method, e.g. the k-TSP and TST, in order to make these methods feasible and cost effective.

Made available in DSpace on 2014-10-09T12:43:30Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:09:43Z (GMT). No. of bitstreams: 1 06636.pdf: 4715244 bytes, checksum: 8df295514eae9493e07c3d6352c4bf32 (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP