In this project, we implement portfolio theory to construct our portfolio, applying the theory to real practice. There are 3 parts in this project, including portfolio optimization, Capital Asset Pricing Model (CAPM) analysis and Factor Model analysis. We implement portfolio theory in the portfolio optimization part. In the second part, we use the CAPM to analyze and improve our portfolio. In the third part we extend our CAPM to factor models to get a deeper analysis of our portfolio.

<p>My thesis research has centered on the concept of the peripheral blood cell (PBC) as an indicator of disease and environmental exposure. The PBC is not only easily accessible and constantly replenished, but it provides a snapshot of an individual's health. Doctors have long utilized PBCs as indicators of health based on count, morphology or the expression of particular cell surface markers. Using these methods, PBCs can serve as indicators of infection, inflammation or certain types of hematological malignancies. Now PBCs can be characterized as a function of their gene expression profiles in response to disease and toxicant exposure. Advances in cDNA microarray technology have made it possible to analyze global gene expression in small volumes of whole blood, or even in a sorted population of blood cells. The resulting gene expression data can serve as a molecular phenotype, or signature, of disease or toxicant exposure. These signatures serve a twofold purpose. First, they act as biological markers (biomarkers) that can indicate the presence of disease or aid monitoring the response to treatment. Second, they provide insight into the underlying biological mechanisms that are at work, by revealing genes, networks and pathways that are affected by the disease or toxin. This paradigm has been applied in a number of contexts, including infection, inflammation, leukemia, lymphoma, neurological disorders, cardiovascular disease, environmental exposures and solid tumors. </p><p>In the work presented here, we describe signatures of lead (Pb) exposure and breast cancer based on peripheral blood gene expression. Our objective in generating a blood-based signature of lead exposure was to develop a potential predictor of past and present exposure. This is particularly relevant because of continued widespread lead exposure through both environmental and occupational sources. Pb causes significant toxicities in a number of different organ systems including the hematological, endocrine, neurological and renal systems. Pb is considered a potential carcinogen due to evidence that it causes cancer in animal models and contributes to an elevated cancer risk in humans. Pb is thought to contribute to cancer risk indirectly through a variety of mechanisms, such as inhibition of DNA synthesis and repair, oxidative damage, interaction with DNA-binding proteins and tumor suppressor proteins, causing chromosomal aberrations and alterations to gene transcription. In addition, it has been shown to exacerbate the effects of other mutagens. Recent work also indicates that even low-level Pb exposure (defined here as levels below the threshold of detection of many common tests or below the level set by the CDC as an "elevated blood lead level" in children, or 10µg/dL) can impact health, especially in children, who are more susceptible to these negative health consequences. </p><p>We hypothesized that we could detect subtle and lasting changes in the PBC transcriptome that correlated to Pb exposure. We used a mouse model of per os Pb exposure to generate signatures corresponding to two different doses of Pb. One dose reflected a high-level exposure and the other a low-level exposure. We also analyzed the gene expression changes following removal of the Pb source. We were able to generate robust, dose-specific signatures of Pb exposure. This supports the growing body of evidence that even low levels of Pb exposure can have biological effects, and that there is likely no safe level of exposure. We also utilized a collection of pathway signatures to identify those pathways that were activated or repressed in response to Pb exposure compared to controls. We observed an increase in interferon-gamma pathway activity in response to low-level Pb exposure and an increase in E2F1 pathway activity in response to high-level Pb exposure. These results support previous findings that low-level Pb exposure can increase interferon-gamma production, whereas high-level Pb has been shown to increase DNA synthesis. The Pb signatures we report here were not predictive of a past lead exposure. These results suggest that the effect of Pb exposure on PBC gene expression is transient, perhaps due to the rapid turnover of blood cells and the absorption of Pb by the bones. We have proposed further studies to identify cells in the bone marrow that may serve as indicators of past Pb exposure based previous reports on the lasting effects of genotoxic stress on this tissue.</p><p>We also describe a predictor of human breast cancer based on peripheral blood gene expression. The objective of this study was to identify and characterize PBC gene expression patterns associated with the presence of a breast tumor. This work has the potential to make a significant impact on breast cancer screening and diagnosis. Despite the success of mammography in reducing mortality from breast cancer, many cancers go undetected due to factors such as breast density, age of the woman, or type of cancer. A blood-based breast tumor predictor would potentially offer an easy and noninvasive means of detecting primary breast cancer as well as monitoring patients for recurrences or metastases. In addition, the concept of using a blood-based biomarker for cancer detection would have positive implications for other types of cancer. For instance, patients with ovarian cancer are typically diagnosed at a late stage because of the absence of definitive symptoms and the lack of effective screenings methods. </p><p>We were able to successfully identify robust predictors of both mouse mammary tumors and human breast tumors based on PBC gene expression. The human breast tumor predictor exhibits a high level of sensitivity and specificity in distinguishing breast cancer patients and controls in an independent validation cohort. However, the true novelty in this study is that it integrates a factor modeling approach and a transgenic mouse model of breast cancer to identify biologically meaningful gene expression changes in the mouse PBC transcriptome. These genes were then used as the starting point for developing a human breast cancer predictor. This establishes an experimental system in which we can address questions that are inherently difficult to answer in human studies, such as whether this predictor is useful in detecting breast tumors early or in monitoring patients for recurrence or metastasis. In fact, our work suggests that tumor-associated gene expression changes in the PBCs can be detected in asymptomatic mice. Our results support those of previous studies, which identified blood gene expression profiles that are associated with a variety of solid tumors, including breast cancer. However, the sensitivity and specificity of our predictor are higher than that of the previously reported breast cancer signature. This may suggest that our strategy of using a mouse model to first identify informative genes allowed us to focus on those genes most relevant to the presence of a breast tumor and overcome the influence of the high degree of variation in blood gene expression in our human population. In order to be clinically useful, the predictor we report here would need to be tested in additional, large validation sets to establish its utility in an early detection setting and its specificity in distinguishing breast cancer from other cancer types as well as other potentially confounding conditions such as infection and inflammation. We describe some preliminary experiments in the mouse model intended to address these important questions.</p> / Dissertation

<div>This dissertation consists of two parts. The first part is a collaborative project with Dr. Szymanski's group in Agronomy at Purdue, to predict protein complex assemblies and interactions. Proteins in the leaf cytosol of Arabidopsis were fractionated using Size Exclusion Chromatography (SEC) and mixed-bed Ion Exchange Chromatography (IEX).</div><div>Protein mass spectrometry data were obtained for the two platforms of separation and two replicates of each. We combine the four data sets and conduct a series of statistical learning, including 1) data filtering, 2) a two-round hierarchical clustering to integrate multiple data types, 3) validation of clustering based on known protein complexes,</div><div>4) mining dendrogram trees for prediction of protein complexes. Our method is developed for integrative analysis of different data types and it eliminates the difficulty of choosing an appropriate cluster number in clustering analysis. It provides a statistical learning tool to globally analyze the oligomerization state of a system of protein complexes.</div><div><br></div><div><br></div><div>The second part examines global hypothesis testing under sparse alternatives and arbitrarily strong dependence. Global tests are used to aggregate information and reduce the burden of multiple testing. A common situation in modern data analysis is that variables with nonzero effects are sparse. The minimum p-value and higher criticism tests are particularly effective and more powerful than the F test under sparse alternatives. This is the common setting in genome-wide association study (GWAS) data. However, arbitrarily strong dependence among variables poses a great challenge towards the p-value calculation of these optimal tests. We develop a latent variable adjusted method to correct minimum p-value test. After adjustment, test statistics become weakly dependent and the corresponding null distributions are valid. We show that if the latent variable is not related to the response variable, power can be improved. Simulation studies show that our method is more powerful than other methods in highly sparse signal and correlated marginal tests setting. We also show its application in a real dataset.</div>

Целью исследования является разработка усовершенствованной методики оценки экономической безопасности предприятий нефтегазовой отрасли. Поставленная цель достигается посредством решения следующего ряда задач: рассмотреть теоретические и методологические основы оценки экономической безопасности; разработать усовершенствованную методику оценки, учитывающую особенность и спецификацию предприятия нефтегазового комплекса; рассчитать по усовершенствованной методики оценку экономической безопасности ПАО «Газпром»; выявить направления по укреплению экономической безопасности ПАО «Газпром. Научная новизна заключается в следующем: 1. Представлено и раскрыто уточненное определение понятия "Экономическая безопасность предприятия НГО", призванное стать теоретической основой формирования предмета оценочных мероприятий в рамках комплексных аналитических процедур. 2. Усовершенствована методика анализа экономической безопасности предприятий НГО, адаптированная к специфике деятельности предприятий НГО и представляющая возможность формирования интегрированного оценочного суждения об уровне экономической безопасности объекта исследования. 3. Разработан регламент внедрения системы оценки экономической без-опасности, призванный сформировать организационную основу управления экономической безопасностью предприятий НГО. / The aim of the study is to develop an improved methodology for assessing the economic security of the oil and gas industry. The goal is achieved through the solution of the following series of tasks: to consider the theoretical and methodological basis for assessing economic security; to develop an improved evaluation methodology that takes into account the peculiarities and specifications of the oil and gas complex; to calculate, based on the improved methodology, an assessment of the economic security of PJSC Gazprom; Identify areas for strengthening economic security of PJSC Gazprom. Scientific novelty consists in the following: 1. The updated definition of the concept of "Economic security of an OGI enterprise" is presented and disclosed, which is intended to become the theoretical basis for the formation of the subject of evaluation activities within the framework of complex analytical procedures. 2. The methodology for analyzing the economic security of enterprises of the OGI, adapted to the specifics of the activities of enterprises of the OGI, is presented and it is possible to form an integrated assessment of the level of economic safety of the research object. 3. The regulations for the implementation of the economic security assessment system, designed to form the organizational basis for managing the economic security of OGI enterprises, have been developed.

By issuing interest rate derivative contracts, market makers such as large banks are exposed to undesired risk. There are several methods for banks to hedge themselves against this type of risk; one such method is the stochastic programming model developed by Blomvall and Hagenbjörk (2022). The effectiveness of their model relies on accurate pricing of interest rate derivatives and risk factor analysis, both of which are derived from a term structure. Blomvall and Ndengo (2013) present a discretized multiple yield curve framework for term structure measurement that allows for price deviations. The model uses regularization to deal with noise inherent in market price observations, where the regularization counteracts oscillations in the term structure and retains the smoothness of the curve by penalizing the first and second-order derivatives. Consequently, the resulting model creates a trade-off between a smooth curve and market price deviations. Changes in policy rates adjusted by a country’s central bank significantly impact the financial market and its actors. In this thesis, the model developed by Blomvall and Ndengo (2013) was further extended to include these steps in conjunction with monetary policy meetings. Two models were developed to realize the steps in the risk-free curve. The first model introduced an additional deviation term to allow for a shift in the curve. In the second model, the weights in the regularization were adjusted to allow for rapid changes on days surrounding the closest monetary policy meeting. A statistical test was conducted to determine the performance of the two models. The test showed that the model with adjusted regularization outperformed the model with an additional deviation term as well as a benchmark model without steps. However, both step models managed to reduce in-sample pricing errors, while the model with an additional deviation term performed worse than the benchmark model for out-of-sample data, given the current parameter setting. Other parameter combinations would potentially result in different outcomes, but it remains conjectural.