The dependence between multiple lines of business has an important impact on determining loss reserves and risk capital, which are crucial elements of risk management for an insurance portfolio. In this work, we show that the Sarmanov family of multivariate distribution can be used for dependent lines of business using a rank-based method estimation. In fact, an inadequate choice of the dependence structure may negatively impact the estimation of the marginals, which might lead to an undesirable effect on reserve computation. Thus, we propose a two-stage inference strategy in this thesis. We show that this strategy leads to robust estimation and better capture the dependence between the risks. We also show that it leads to smaller risk capital and a better diversification benefit. We introduce the two-stage inference using the Sarmanov distribution. First, we fit the marginals with generalized linear models (GLMs) and obtain the corresponding residuals. Secondly, the Sarmanov family of bivariate distributions links these marginals through the rank of residuals. We also show that this can be extended to a multivariate case. To illustrate this method, we analyzed two sets of data. For the bivariate case, we considered an insurance portfolio consisting of personal and commercial auto lines provided by a major US property-casualty insurer. We also used the data from three lines of business of a large Canadian insurance company for the multivariate dependence case. / Thesis / Master of Science (MSc)

For a graph G we define S(G) to be the set of all real symmetric n by n matrices whose off-diagonal zero/nonzero pattern is described by G. We show how to compute the minimum rank of all matrices in S(G) for a class of graphs called outerplanar graphs. In addition, we obtain results on the possible eigenvalues and possible inertias of matrices in S(G) for certain classes of graph G. We also obtain results concerning the relationship between two graph parameters, the zero forcing number and the path cover number, related to the minimum rank problem.

Épreuves Classantes Nationales informatisés (ECNi) är ett nationellt prov som franska läkarstudenter skriver för att gå vidare till sin tredje utbildningsperiod, där de väljer sin specialitet och region. Studenter som presterar bäst på provet får jobberbjudanden inom den region som är deras preferens. Det är därför viktigt för studenterna att veta deras relativa eller estimera de placering innan de genomför provet. Hypocampus är en adaptiv webbaserad läroplattform vilket används av franska läkarstudenter för att förbereda sig för ECNi-provet. Plattformen erbjuder personlig anpassningsbar utbildningsplan utifrån kognitiv förmåga, kunskapsnivå och visualiserar inlärningsprocessen för studenter och lärare. Dock, erbjuder inte plattformen möjligheten för studenter att jämföra sin egna prestation mot andra studenter. Syftet med denna studie är därför att hitta en visualiseringsteknik för att jämföra studenters prestation. Flera visualiseringstekniker valdes och utvärderades med hjälp av kvalitativa och kvantitativa metoder. Resultatet från studien visar att det finns lovande resultat inom att använda datavisualisering som tillvägagångssätt vid jämförande av studenters prestation och att visualisering med histogram föredrogs över andra visualiseringar inom kontexten att jämföra en students prestation mot andra studenter i systemet. / The Épreuves Classantes Nationales informatisés (ECNi) is a national ranking examination that all medical students in France must pass to proceed to the third training cycle, where they choose their specialty and medical region. Students who achieve the highest rankings will be offered jobs in the geographic location of their choice. It is, therefore, important that students know their relative or estimated position on the ranked list before they take the exam. Hypocampus is an online adaptive web-based learning platform used by medical students in France to prepare for the ECNi. The platform offers a personalized learning path adaptable to students’ cognitive skills, knowledge level, and visualizations of learning progress for students and teachers. However, it does not allow students to compare their performance or knowledge with other students. This work aims to explore different information visualization techniques for visual comparison of students’ performance. Several visualization prototypes were designed and evaluated with the expert reviews method. The findings of this work show promising results in using the data visualization approach for comparison of students’ performance and that histogram-visualization was preferred over the other visualizations in the context of comparing a students’ performance to other students in the system.

The main goal of this thesis is to investigate query-dependent selection of retrieval alternatives for Information Retrieval (IR) systems. Retrieval alternatives include choices in representing queries (query representations), and choices in methods used for scoring documents. For example, an IR system can represent a user query without any modification, automatically expand it to include more terms, or reduce it by dropping some terms. The main motivation for this work is that no single query representation or retrieval model performs the best for all queries. This suggests that selecting the best representation or retrieval model for each query can yield improved performance. The key research question in selecting between alternatives is how to estimate the performance of the different alternatives. We treat query dependent selection as a general problem of selecting between the result sets of different alternatives. We develop a relative effectiveness estimation technique using retrieval-based features and a learning formulation that directly predict differences between the results sets. The main idea behind this technique is to aggregate the scores and features used for retrieval (retrieval-based features) as evidence towards the effectiveness of the results set. We apply this general technique to select between alternatives reduced versions for long queries and to combine multiple ranking algorithms. Then, we investigate the extension of query-dependent selection under specific efficiency constraints. Specifically, we consider the black-box meta-search scenario, where querying all available search engines can be expensive and the features and scores used by the search engines are not available. We develop easy-to-compute features based on the results page alone to predict when querying an alternate search engine can be useful. Finally, we present an analysis of selection performance to better understand when query-dependent selection can be useful.

The Oshawa 1937 strike against General Motors was a major turning point in Canadian labour history. This thesis explores the factors that led to its success, including the historical background of working class struggle; the economic and political context of the times; prior organizing by Communists; the engagement of rank-and-file GM workers and the remarkable stewards’ body they established; and the support and leadership of the UAW International union. The thesis shows there was overlap and interplay between these factors. The influence of the strategic outlook of Communists, both in Oshawa and in the UAW more broadly, meant that the 1937 strike incorporated many features of what might now be called rank-and-file unionism: industrial unionism, democratic engagement of rank-and-file workers, militancy on the shop floor, building solidarity within the workforce and in the community, international solidarity, and rejecting cooperation with corporations. My research focus was on the voices and actions of rank-and-file workers as much as possible, and on the remarkable day-to-day events of the strike itself. The thesis demonstrates that many of the events and lessons of the strike have been little understood or have been misinterpreted. In particular, I reexamine and correct the long-accepted conclusions of Irving Abella that the Oshawa workers were “on their own” without significant support from the UAW/CIO leadership, and that they (or the Canadian labour movement) would have been better off if they did not organize under the banner of an international union. The thesis also demonstrates that Abella failed to grasp the degree to which rank-and-file principles were embodied by the strike and were crucial to its success. The contending forces of workers, corporations, and rabidly anti-union governments that clashed in Oshawa in 1937 are largely the same ones we see in the battles going on in North America today. Thus, understanding the factors that led to the success of the Oshawa strikers can provide valuable lessons to those seeking to revive today’s labour movement. / Thesis / Master of Arts (MA) / The Oshawa 1937 strike against General Motors was a major turning point in Canadian labour history. This thesis explores the factors that led to its success, including the historical background of working class struggle; the economic and political context of the times; prior organizing by Communists; the engagement of rank-and-file GM workers and the remarkable stewards’ body they established; and the support and leadership of the UAW International union. The influence of Communists meant that the strike incorporated many features of what might now be called rank-and-file unionism: industrial unionism, democratic engagement of rank-and-file workers, militancy on the shop floor, building solidarity within the workforce and in the community, international solidarity, and rejecting cooperation with corporations. The contending forces of workers, corporations, and rabidly anti-union governments that clashed in Oshawa in 1937 are largely the same ones we see in the battles going on in North America today. Thus, understanding the factors that led to the success of the Oshawa strikers can provide valuable lessons to those seeking to revive today’s labour movement.

The intent of this thesis is to explore whether any patterns emerge among families or through graph operations regarding the appearance of Weierstrass vertices on graphs. Currently, patterns have been identified and proven on cycles, complete graphs, complete bipartite graphs, and the house and house-x graphs. A Python program developed as part of this thesis to perform the algorithms used in this analysis confirms these findings. This program also revealed a pattern: if v is a Weierstrass vertex, then the vertex v* added to the graph as a pendant vertex to v is also a Weierstrass vertex. The converse is also true: if v is not a Weierstrass vertex, v* will not be either.

No description available.

No description available.

No description available.

The practice of ranking objects, events, and people to determine relevance, importance, or competitive edge is ancient. Recently, the use of rankings has permeated into daily usage, especially in the fields of business and education. When determining the association among those creating the ranks (herein called sources), the traditional assumption is that all sources compare a list of the same items (herein called elements). In the twenty-first century, it is rare that any two sources choose identical elements to rank. Adding to this difficulty, the number of credible sources creating and releasing rankings is increasing. In statistical literature, there is no current methodology that adequately assesses the association among multiple sources. We introduce rank-adapted singular value decomposition (R-A SVD), a new method that uses Kendall's tau as the underlying correlation method. We begin with (P), a matrix of data ranks. The first step is to factor the covariance matrix (K) as follows: K = cov(P) = V D^2 V Here, (V) is an orthonormal basis for the rows that is useful in identifying when sources agree as to the rank order and specifically which sources. D is a diagonal of eigenvalues. By analogy with singular value decomposition (SVD), we define U^* as U^* = PVD^(-1) The diagonal matrix, D, provides the factored eigenvalues in decreasing order. The largest eigenvalue is used to assess the overall association among the sources and is a conservative unbiased method comparable to Kendall's W. Anderson's test determines whether this association is significant and also identifies other significant eigenvalues produced by the covariance matrix.. Using Anderson's test (1963) we identify the a significantly large eigenvalues from D. When one or more eigenvalues is significant, there is evidence that the association among the sources is significant. Focusing on the a corresponding vectors of V specifically identifies which sources agree. In cases where more than one eigenvalue is significant, the $a$ significant vectors of V provide insight into factions. When more than one set of sources is in agreement, each group of agreeing sources is considered a faction. In many cases, more than one set of sources will be in agreement with one another but not necessarily with another set of sources; each group that is in agreement would be considered a faction. Using the a significant vectors of U^* provides different but equally important results. In many cases, the elements that are being ranked can be subdivided into equivalence classes. An equivalence class is defined as subpopulations of ranked elements that are similar to one another but dissimilar from other classes. When these classes exist, U^* provides insight as to how many classes and which elements belong in each class. In summary, the R-A SVD method gives the user the ability to assess whether there is any underlying association among multiple rank sources. It then identifies when sources agree and allows for more useful and careful interpretation when analyzing rank data. / Statistics