Empirical likelihood, which was pioneered by Thomas and Grunkemeier (1975) and Owen (1988), is a powerful nonparametric method of statistical inference that has been widely used in the statistical literature. In this thesis, we investigate the merits of empirical likelihood for various problems arising in ratio estimation. First, motivated by the smooth empirical likelihood (SEL) approach proposed by Zhou & Jing (2003), we develop empirical likelihood estimators for diagnostic test likelihood ratios (DLRs), and derive the asymptotic distributions for suitable likelihood ratio statistics under certain regularity conditions. To skirt the bandwidth selection problem that arises in smooth estimation, we propose an empirical likelihood estimator for the same DLRs that is based on non-smooth estimating equations (NEL). Via simulation studies, we compare the statistical properties of these empirical likelihood estimators (SEL, NEL) to certain natural competitors, and identify situations in which SEL and NEL provide superior estimation capabilities. Next, we focus on deriving an empirical likelihood estimator of a baseline cumulative hazard ratio with respect to covariate adjustments under two nonproportional hazard model assumptions. Under typical regularity conditions, we show that suitable empirical likelihood ratio statistics each converge in distribution to a 2 random variable. Through simulation studies, we investigate the advantages of this empirical likelihood approach compared to use of the usual normal approximation. Two examples from previously published clinical studies illustrate the use of the empirical likelihood methods we have described. Empirical likelihood has obvious appeal in deriving point and interval estimators for time-to-event data. However, when we use this method and its asymptotic critical value to construct simultaneous confidence bands for survival or cumulative hazard functions, it typically necessitates very large sample sizes to achieve reliable coverage accuracy. We propose using a bootstrap method to recalibrate the critical value of the sampling distribution of the sample log-likelihood ratios. Via simulation studies, we compare our EL-based bootstrap estimator for the survival function with EL-HW and EL-EP bands proposed by Hollander et al. (1997) and apply this method to obtain a simultaneous confidence band for the cumulative hazard ratios in the two clinical studies that we mentioned above. While copulas have been a popular statistical tool for modeling dependent data in recent years, selecting a parametric copula is a nontrivial task that may lead to model misspecification because different copula families involve different correlation structures. This observation motivates us to use empirical likelihood to estimate a copula nonparametrically. With this EL-based estimator of a copula, we derive a goodness-of-fit test for assessing a specific parametric copula model. By means of simulations, we demonstrate the merits of our EL-based testing procedure. We demonstrate this method using the data from Wieand et al. (1989). In the final chapter of the thesis, we provide a brief introduction to several areas for future research involving the empirical likelihood approach.

Le paludisme est un problème de santé public dans de nombreux pays. Cinq espèces infectent l'homme : P. falciparum, responsable de la grande majorité des décès, et P. vivax, P. ovale, P. malariae et P. knowlesi qui provoquent des formes bénignes de la maladie. Le diagnostic qui fait partie des moyens de lutte, est une urgence médicale. Les tests de diagnostic rapides (TDRs) dont l'usage est recommandés par l'OMS, sont donc de plus en plus employés. Cependant, la détection et l'identification des espèces non P. falciparum par ces tests est insuffisante. Le besoin en nouveaux couples « antigènes-anticorps » est une nécessité pour améliorer les TDRs. Au cours de ce travail, de nouveaux anticorps anti LDH de P.malariae ont été produits.Une recherche de nouveaux antigènes a également été entreprise. Pour cela, certaines enzymes de la voie de la glycolyse ont été étudiées. Pour la première fois des séquences des enzymes de cette voie ont été obtenues pour P. ovale et P. malariae. Elles ont permis de déterminer de nombreux épitopes cibles potentiels spécifiques et ceux communs à toutes les espèces. Dans un deuxième temps, une recherche en protéomique a été menée pour identifier des biomarqueurs parasitaires. L'étude du culot globulaire et du plasma de patients infectés a permis la sélection de 8 protéines cibles originales. Ces travaux préparent la fabrication et la commercialisation par la société Whidiag d'une nouvelle génération de TDRs pour le paludisme. / Malaria is a public health problem in many countries. Five species infect humans: P. falciparum, responsible for the vast majority of deaths, and P. vivax, P. ovale, P. malariae and P. knowlesi causing mild forms of the disease. The diagnostic is a means of control and a medical emergency. The rapid diagnostic tests (RDT) whose are recommended by WHO, are increasingly used. However, the detection and identification of not P. falciparum species is insufficient. New "antigen-antibody" couples are a need to improve the RDTs performance. In this work, new anti LDH antibodies from P. malariae were produced. A search for new antigens was also undertaken. For this purpose, some enzyme of glycolysis pathway were studied. For the first time the sequences of the enzymes from this pathway were obtained for P. ovale and P. malariae. We identified many potential target epitopes specific and common to all those species. In a second step, a proteomics approches has been conducted to identify parasites biomarkers. The study of red blood cells and plasma of infected patients has led to the selection of 8 original target proteins. This work prepares the manufacturing and marketing of a new generation of RDTs for malaria by the company Whidiag

TITLE: The Lexicon Evaluation of Children before their Enrolling in School AUTHOR: Lucie Durdilová DEPARTMENT: Special Education SUPERVISOR: Doc. PaedDr. Jiřina Klenková, Ph.D. Abstract: The submitted thesis is focused on evaluation of lexical - semantic levels of language in children in late pre-school age, it specifically follows up their achievement in parts of receptive and expressive vocabulary. The measurement is implemented through the diagnostic test method which was compiled on the basis of foreign inspiration drawn especially from anglophone linguistic ambience. The lexicon is measured in intact population and in children with specific language impairment with the aim to compare obtained data. Developed findings should help to increase efficiency of logopedic intervention in content part of the speech. The project is based on the fact that currently there are no useful update tests available in the Czech Republic for global evaluation of lexicon. Keywords: receptive vocabulary, expressive vocabulary, diagnostic test, pre-school age

A majority of students at the local University College of Science and Education (UCSE, pseudonym) in Jamaica do not have the conceptual understanding of mathematical principles to function in a competitive and highly globalized marketplace. In 2013 and 2014, 88% and 92% of freshmen education students scored at the lowest 2 levels on the Mathematics Diagnostic Test (MDT). The instructional language at UCSE is Standard English (SE) whereas most students speak Jamaican dialect (JD). The purpose of this study was to determine the effect that the language of instruction has on student achievement in math as measured by the MDT. Guided by Vygotsky's social development theory, the research questions focused on comparing MDT change scores between students who were taught using JD and those using SE as the instructional language. The quasi-experimental design used ex post facto data including pretest and posttest MDT scores from 40 freshmen of whom 20 were instructed in JD and 20 in SE. The results of an independent sample t test showed that the difference in the MDT change score was significant. The JD students had a higher improvement score. Consequently, it is recommended that math instructors use JD to instruct freshmen education students whose native language is JD. A professional development session for math teachers was created that demonstrates how to teach in JD while simultaneously scaffolding the instruction in a way that students can learn SE and be prepared for the following year at UCSE. If students understand the math concepts in their freshman year, they are more likely to continue their college education and to become productive members of Jamaica's economy which is dependent on employees that are proficient in math.

I detta examensarbete har en intervjubaserad undersökning genomförts, med tio gymnasieelever som studerar på det naturvetenskapliga programmet. Syftet var att ta reda på elevernas perspektiv på hur deras eget lärande i kemi kopplas till diagnostiska test, samt i vilka kontexter eleverna nämner diagnostiska test. Överlag verkar eleverna i undersökningen ha god förståelse av vad syftet är med diagnostiska test, men de tar frekvent upp negativa känslor som stress, prestationsångest och oro över om diagnostiska test påverkar betyget. Eleverna tog upp temat rättvisa i relationen mellan undervisning, lärarens förhållningssätt och bedömning av diagnostiska test. Teman som lärande och lektionsplanering är genomgående i elevernas svar där lärarens situation tas i beaktande. Det gäller inte minst för att planera och anpassa undervisningen till både klassen och enskilda elevers behov. / In this thesis, an interview-based survey has been conducted, comprising of ten high school students studying the science program. The purpose was to find out the students' perspectives on how their own learning in chemistry is linked to diagnostic tests, and in which contexts the students mention diagnostic tests. Overall high school students seem to have a good understanding about the purpose of diagnostic tests, but frequently address negative emotions such as stress, performance anxiety and concerns about whether diagnostic tests affect the grade. The theme of fairness arose where students discussed the relationship between teaching, the teacher's conduct and the assessment of diagnostic tests. Learning and lesson planning are recurrent themes in students' responses where the teacher's situation is taken into account. This applies not least to planning and adapting the teaching to both the class and individual students.

Meta-analysis is a popular statistical method that synthesizes evidence from multiple studies. Conventionally, both the hierarchical and bivariate models for meta-analysis of diagnostic test accuracy (DTA) studies assume that the random-effects follow the bivariate normal distribution. However, this assumption is restrictive, and inferences could be misleading when it is violated. On the other hand, subjective methods such as inspection of forest plots are used to identify outlying studies in a meta-analysis of DTA studies. Moreover, inferences made using the well-established bivariate random-effects models, when outlying or influential studies are present, may lead to misleading conclusions. Thus, the aim of this thesis is to address these issues by introducing alternative and robust statistical methods. First, we extend the current bivariate linear mixed model (LMM) by assuming a flexible bivariate skew-normal distribution for the random-effects. The marginal distribution of the proposed model is analytically derived so that parameter estimation can be performed using standard likelihood methods. Overall, the proposed model performs better in terms of confidence interval width of the overall sensitivity and specificity, and with regards to bias and root mean squared error of the between-study (co)variances than the traditional bivariate LMM. Second, we propose objective methods based on solid statistical reasoning for identifying outlying and/or influential studies in a meta-analysis of DTA studies. The performances of the proposed methods are evaluated using a simulation study. The proposed methods outperform and avoid the subjectivity of the currently used ad hoc approaches. Finally, we develop a new robust bivariate random-effects model which accommodates outlying and influential observations and leads to a robust statistical inference by down-weighting the effect of outlying and influential studies. The proposed model produces robust point estimates of sensitivity and specificity compared to the standard models, and also generates a similar point and interval estimates of sensitivity and specificity as the standard models in the absence of outlying or influential studies. / Thesis / Doctor of Philosophy (PhD) / Diagnostic tests vary from the noninvasive rapid strep test used to identify whether a patient has a bacterial sore throat to the much complex and invasive biopsy test used to examine the presence, cause, and extent of a severe condition, say cancer. Meta-analysis is a widely used statistical method that synthesizes evidence from several studies. In this thesis, we develop novel statistical methods extending the traditional methods for meta-analysis of diagnostic test accuracy studies. Our proposed methods address the issue of modelling asymmetrical data, identifying outlier studies, and optimally accommodating these outlying studies in a meta-analysis of diagnostic test accuracy studies. Using both real-life and simulated datasets, we show that our proposed methods perform better than conventional methods in a wide range of scenarios. %Therefore, we believe that our proposed methods are essential for methodologists, clinicians and health policy professionals in the process of making a correct judgment to using the appropriate diagnostic test to diagnose patients.

The test process for semiconductor devices involves generation and application of test patterns, failure logging and diagnosis. Traditionally, most of these activities cater for all possible faults without making any assumptions about the actual defects present in the circuit. As the size of the circuits continues to increase (following the Moore's Law) the size of the test sets is also increasing exponentially. It follows that the cost of testing has already surpassed that of design and fabrication. The central idea of our work in this dissertation is that we can have substantial savings in the test cost if we bring the actual hardware under test inside the test process's various loops -- in particular: failure logging, diagnostic pattern generation and diagnosis. Our first work, which we describe in Chapter 3, applies this idea to failure logging. We modify the existing failure logging process that logs only the first few failure observations to an intelligent one that logs failures on the basis of their usefulness for diagnosis. To enable the intelligent logging, we propose some lightweight metrics that can be computed in real-time to grade the diagnosibility of the observed failures. On the basis of this grading, we select the failures to be logged dynamically according to the actual defects in the circuit under test. This means that the failures may be logged in a different manner for devices having different defects. This is in contrast with the existing method that has the same logging scheme for all failing devices. With the failing devices in the loop, we are able to optimize the failure log in accordance with every particular failing device thereby improving the quality of diagnosis subsequently. In Chapter 4, we investigate the most lightweight of these metrics for failure log optimization for the diagnosis of multiple simultaneous faults and provide the results of our experiments. Often, in spite of exploiting the entire potential of a test set, we might not be able to meet our diagnosis goals. This is because the manufacturing tests are generated to meet the fault coverage goals using as fewer tests as possible. In other words, they are optimized for `detection count' and `test time' and not for `diagnosis'. In our second work, we leverage realtime measures of diagnosibility, similar to the ones that were used for failure log optimization, to generate additional diagnostic patterns. These additional patterns help diagnose the existing failures beyond the power of existing tests. Again, since the failing device is inside the test generation loop, we obtain highly specific tests for each failing device that are optimized for its diagnosis. Using our proposed framework, we are able to diagnose devices better and faster than the state of the art industrial tools. Chapter 5 provides a detailed description of this method. Our third work extends the hardware-in-the-loop framework to the diagnosis of scan chains. In this method, we define a different metric that is applicable to scan chain diagnosis. Again, this method provides additional tests that are specific to the diagnosis of the particular scan chain defects in individual devices. We achieve two further advantages in this approach as compared to the online diagnostic pattern generator for logic diagnosis. Firstly, we do not need a known good device for generating or knowing the good response and secondly, besides the generation of additional tests, we also perform the final diagnosis online i.e. on the tester during test application. We explain this in detail in Chapter 6. In our research, we observe that feedback from a device is very useful for enhancing the quality of root-cause investigations of the failures in its logic and test circuitry i.e. the scan chains. This leads to the question whether some primitive signals from the devices can be indicative of the fault coverage of the applied tests. In other words, can we estimate the fault coverage without the costly activities of fault modeling and simulation? By conducting further research into this problem, we found that the entropy measurements at the circuit outputs do indeed have a high correlation with the fault coverage and can also be used to estimate it with a good accuracy. We find that these predictions are accurate not only for random tests but also for the high coverage ATPG generated tests. We present the details of our fourth contribution in Chapter 7. This work is of significant importance because it suggests that high coverage tests can be learned by continuously applying random test patterns to the hardware and using the measured entropy as a reward function. We believe that this lays down a foundation for further research into gate-level sequential test generation, which is currently intractable for industrial scale circuits with the existing techniques. / Ph. D. / When a new microchip fabrication technology is introduced, the manufacturing is far from perfect. A lot of work goes into updating the fabrication rules and microchip designs before we get a higher proportion of good or defect-free chips. With continued advancements in the fabrication technology, this enhancement work has become increasingly difficult. This is primarily because of the sheer number of transistors that can be fabricated on a single chip, which has practically doubled every two years for the last four decades. The microchip testing process involves application of stimuli and checking the responses. These stimuli cater for a huge number of possible defects inside the chips. With the increase in the number of transistors, covering all possible defects is becoming practically impossible within the business constraints. This research proposes a solution to this problem, which is to make various activities in this process adaptive to the actual defects in the chips. The stimuli, we mentioned above, now depend upon the feedback from the chip. By utilizing this feedback, we have demonstrated significant improvements in three primary activities namely failure logging, scan testing and scan chain diagnosis over state-of-the-art industrial tools. These activities are essential steps related to improving the proportion of good chips in the manufactured lot.

La scoliose idiopathique de l’adolescent (SIA) est une maladie dont la cause est encore inconnue, et qui génère des déformations complexes du rachis, du thorax et du bassin. La prévalence est de 4% dans la population adolescente au Québec. Cette pathologie affecte surtout les filles durant leur poussée de croissance pubertaire. Parmi plusieurs hypothèses émises, l’hypothèse neuroendocrinienne, impliquant une déficience en mélatonine comme agent étiologique de la SIA a suscité beaucoup d’intérêt. Cette hypothèse découle du fait que l’ablation de la glande pinéale chez le poulet produit une scoliose ressemblant sous plusieurs aspects à la pathologie humaine. La pertinence biologique de la mélatonine dans la scoliose est controversée, étant donné que la majorité des études chez l’homme n’ont pu mettre en évidence une diminution significative des niveaux de mélatonine circulante chez les patients scoliotiques. Nous avons démontré un dysfonctionnement dans la signalisation de la mélatonine au niveau des tissus musculo-squelettiques chez une série de patients atteints de SIA (Moreau & coll. 2004). Nous avons confirmé ce défaut chez un plus grand nombre de patients ainsi qu’en utilisant une nouvelle technologie (spectroscopie cellulaire diélectrique) n’ayant pas recours à un prétraitement des cellules donnant ainsi des résultats plus précis. Cette technique a montré la présence des mêmes groupes fonctionnels identifiés auparavant par la technique d’AMPc. Le dysfonctionnement de la signalisation de la mélatonine est dû à une phosphorylation accrue des protéines G inhibitrices. Ce défaut pourrait être causé par un déséquilibre de l’activité des kinases et phosphatases capables de réguler la phosphorylation des protéines Gi. Parmi ces kinases, PKCd a suscité initialement notre intérêt vu qu’elle peut phosphoryler les protéines Gi. Nous avons démontré que cette kinase interagit avec le récepteur de la mélatonine MT2 et que cette interaction varie selon le groupe fonctionnel auquel un patient SIA appartient. Par la suite nos travaux se sont dirigés vers la découverte d’effecteurs cellulaires régulés par la mélatonine et plus spécifiquement l’ostéopontine (OPN), compte tenu de son rôle présumé comme mécanorécepteur et dans certaines structures jouant un rôle dans la proprioception, le contrôle postural et la fonction vestibulaire. L’OPN a été identifiée initialement par sa surexpression au niveau protéique et de l’ARNm dans la musculature paraspinale uniquement chez les poulets scoliotiques. Nous avons également utilisé un autre modèle animal, la souris C57Bl/6 naturellement déficiente en mélatonine. Nous avons généré des souris bipèdes en amputant les membres antérieurs de souris OPN KO, des souris CD44 KO ainsi que des souris contrôles C57Bl/6. Nos résultats ont montré qu’aucune souris OPN KO (n=50) ou CD44 KO (n=60) ne développe la maladie, contrairement aux souris contrôles C57Bl/6 (n=50) dont 45% deviennent scoliotiques. Ces résultats nous ont poussés à investiguer le rôle de cette protéine dans l’étiopathogenèse de la maladie chez l’humain. Nos résultats ont montré une augmentation des niveaux circulants d’OPN chez les patients atteints de la SIA et que l’élevation en OPN corrélait avec la sévérité de la maladie. Nos études chez les enfants asymptomatiques nés de parents scoliotiques et qui sont plus à risque de développer la maladie ont aussi démontré des différences significatives au niveau des concentrations en OPN en comparaison avec les sujets sains. En effet, plusieurs enfants à risque présentaient des niveaux d’OPN supérieurs à 800ng/ml suggérant un plus grand risque de développer une scoliose indiquant aussi que l’augmentation des niveaux en OPN précède le début de la maladie. / Adolescent idiopathic scoliosis (AIS) is the most common form of scoliosis that affects a significant number of young teenagers, mainly females. Historically, several hypotheses were postulated to explain the aetiology of AIS. The neuroendocrine hypothesis involving a melatonin deficiency as the source for AIS has generated great interest. This hypothesis stems from the fact that experimental pinealectomy in chickens, and more recently in rats maintained in a bipedal mode, produces scoliosis. The biological relevance of melatonin in idiopathic scoliosis is controversial since no significant decrease in circulating melatonin level has been observed in a majority of studies. Analysis of melatonin signal transduction in musculoskeletal tissues of AIS patients demonstrated for the first time a defect occurring in a cell autonomous manner in different cell types isolated from AIS patients suffering of the most severe form of that disease. We confirmed this defect by analysing more AIS patients and by using a new technology (cellular dielectric spectroscopy) which gives more precise results because it allows the measurement and analysis of receptor activation without the need to pretreat cells. This technique showed the same functional classification into three functional groups as identified by cAMP technique. Melatonin signalling dysfunction is caused by phosphorylation of serine residues affecting the activity of G inhibitory (Gi) proteins normally associated with melatonin receptors present at the cell surface. This defect could be caused by an imbalance in the activity of kinases or phosphatases that can regulate Gi proteins phosphorylation. Among these kinases PKCd was initially of interest because it has been shown that it can phosphorylate Gi proteins. We showed that this kinase interacts with melatonin receptor MT2 and that this interaction varies from one functional group to another. Thereafter, we moved one step further to characterise downstream effector regulated by melatonin. This work has led to the identification of osteopontin (OPN) which is a relevant candidate because it can act as a mecanosensor and it is involved in proprioception, postural and vestibular control. OPN was initially identified in pinealectomized chickens where it was shown to be upregulated at protein and mRNA levels only in scoliotic ones. We also used another animal model, C57Bl/6 mice which are naturally deficient in melatonin. We generated bipedal mice by amputating forelimbs of OPN knock-out mice, CD44 knock-out mice as well as C57Bl/6 wild type mice. Our results showed that all bipedal mice OPN Knock-out or CD44 Knock-out did not develop scoliosis contrasting with C57Bl/6 wt mice where 45% develop scoliosis. These results prompted us to investigate the role of this protein in scoliosis etiopathogenesis in humans. We showed an increase in the OPN circulating levels in AIS patients and this elevation correlates with disease severity. Elevated plasma OPN levels were also found in the asymptomatic at-risk group (offspring of scoliotic patients), suggesting that these changes precede scoliosis onset.

In the field of diagnostic test studies, the accuracy of a diagnostic test is essential in evaluating the performance of the test. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) are widely used in such evaluation procedures. Meanwhile, the Youden index is also introduced into practice to measure the accuracy of the diagnostic test from another aspect. The Youden index maximizes the sum of sensitivity and specificity, assuring decent true positive and negative rates. It draws one's attention due to its merit of finding the optimal cut-off points of biomarkers. Similar to Partial ROC, a new index, called "Partial Youden index" can be defined as an extension of Youden's Index. It is more meaningful than regular Youden index since the regular one is just a special case of the Partial Youden Index. In this thesis, we focus on construction of generalized confidence intervals for the Partial Youden Index and its corresponding optimal cut-off points. Extensive simulation studies are conducted to evaluate the finite sample performances of the new intervals.

The receiver operating characteristic (ROC) curves is a popular tool for evaluating continuous diagnostic tests. The traditional definition of ROC curves incorporates implicitly the idea of "hard" thresholding, which also results in the empirical curves being step functions. The first topic is to introduce a novel definition of soft ROC curves, which incorporates the idea of "soft" thresholding. The softness of a soft ROC curve is controlled by a regularization parameter that can be selected suitably by a cross-validation procedure. A byproduct of the soft ROC curves is that the corresponding empirical curves are smooth. The second topic is on combination of several diagnostic tests to achieve better diagnostic accuracy. We consider the optimal linear combination that maximizes the area under the receiver operating characteristic curve (AUC); the estimates of the combination's coefficients can be obtained via a non-parametric procedure. However, for estimating the AUC associated with the estimated coefficients, the apparent estimation by re-substitution is too optimistic. To adjust for the upward bias, several methods are proposed. Among them the cross-validation approach is especially advocated, and an approximated cross-validation is developed to reduce the computational cost. Furthermore, these proposed methods can be applied for variable selection to select important diagnostic tests. However, the above best-subset variable selection method is not practical when the number of diagnostic tests is large. The third topic is to further develop a LASSO-type procedure for variable selection. To solve the non-convex maximization problem in the proposed procedure, an efficient algorithm is developed based on soft ROC curves, difference convex programming, and coordinate descent algorithm.