No description available.

The prospect of using the Internet and other Big Data methods to construct event data promises to transform the field but is stymied by the lack of a coherent strategy for addressing the problem of selection. Past studies have shown that event data have significant selection problems. In terms of conventional standards of representativeness, all event data have some unknown level of selection no matter how many sources are included. We summarize recent studies of news selection and outline a strategy for reducing the risks of possible selection bias, including techniques for generating multisource event inventories, estimating larger populations, and controlling for nonrandomness. These build on a relativistic strategy for addressing event selection and the recognition that no event data set can ever be declared completely free of selection bias.

Heterosexual romance and marriage are institutionalized ideals in our society, set forth, in part, through the portrayal of stereotyped gender roles in fairy tales, such as Cinderella, and by the mainstream media. This thesis explores the cultural messages aimed at women, which impose the necessity of altering oneself to achieve marriage, and offers feminist viewpoints. Using the form of the personal essay, I discuss the ideals of Cinderella, Prince Charming, marriage, and Happily Ever After as unrealistic, though still prevalent, given the popularity of books like The Rules: Time-tested Secrets for Capturing the Heart of Mr. Right, and Princess Diana as Cinderella icon. Essays on my own experience of marriage and divorce supplement the cultural issues, juxtaposing the personal and political toward a new paradigm for relationships.

Variations in species occurrence and distribution across the landscape over time provide fundamental information concerning population dynamics. How this relates to habitat characteristics at multiple scales can elucidate the process of habitat selection. I evaluated these processes for a montane fir (Abies) forest specialist, Bicknell’s Thrush (Catharus bicknelli) in Vermont. This species is threatened by a suite of anthropogenic disturbances on its breeding grounds and quantifying the effects of environmental change at the population level for this songbird has not been addressed. The naturally fragmented breeding habitat of varying size, quality, and connectivity warranted a metapopulation approach and a robust occupancy analysis. Detection/non-detection data was collected for Bicknell’s Thrush across 88 sites during the breeding seasons in 2006 and 2007. Local habitat characteristics were measured for each site and landscape-level features were calculated using a predictive habitat model. The six local habitat variables were combined using a principal component analysis. Principal component 1 (PC1) described a gradient of increasing coniferous shrub density and proportion of coniferous dominated forest with decreasing canopy height. The landscape covariate was calculated by dividing patch size by patch isolation creating a continuum of small, isolated patches to large, less isolated patches. Thus each site was characterized by a single local habitat (PC1) and landscape metric. From these data, 67 models considering all combinations of landscape and local habitat scores (univariate, additive and interaction) were evaluated for individual estimates of the following parameters: (1) probability of detection, (2) probability of initial site occupancy, (3) probability of site colonization, and (4) probability of local site extinction. AIC model selection techniques were used to rank the models, which represented ecologically plausible hypotheses that compared the strength of local habitat characteristics to large-scale landscape features. Models within 4 AICc points of the top model were considered plausible. The top eight models were all plausible. Landscape characteristics alone were not significant in driving population dynamics. The relative importance of landscape + local habitat was highest for both probability of initial occupancy and local site extinction. Probability of occupancy increased and extinction decreased with the combination of increased patch size and decreased patch isolation (landscape) and increased coniferous shrub density, proportion of coniferous dominated forest and decreased canopy height (local habitat). Probability of site colonization was driven mainly by local habitat features and increased with increasing habitat quality. These results indicate a complex system with intricate links between landscape and local scales. Preserving large tracts of habitat may not be sufficient in assuring future species persistence, but could minimize local extinction risk. Careful consideration should be given to local habitat features within habitat fragments, particularly to maintain adequate colonization rates. Because important features from both scales are correlated, in intact montane forest patches, landscape-scale attributes alone may serve as a surrogate for identifying quality breeding habitat, assuming processes of natural disturbance can be maintained.

No description available.

Bayesian subset selection suffers from three important difficulties: assigning priors over model space, assigning priors to all components of the regression coefficients vector given a specific model and Bayesian computational efficiency (Chen et al., 1999). These difficulties become more challenging in Bayesian quantile regression framework when one is interested in assigning priors that depend on different quantile levels. The objective of Bayesian quantile regression (BQR), which is a newly proposed tool, is to deal with unknown parameters and model uncertainty in quantile regression (QR). However, Bayesian subset selection in quantile regression models is usually a difficult issue due to the computational challenges and nonavailability of conjugate prior distributions that are dependent on the quantile level. These challenges are rarely addressed via either penalised likelihood function or stochastic search variable selection (SSVS). These methods typically use symmetric prior distributions for regression coefficients, such as the Gaussian and Laplace, which may be suitable for median regression. However, an extreme quantile regression should have different regression coefficients from the median regression, and thus the priors for quantile regression coefficients should depend on quantiles. This thesis focuses on three challenges: assigning standard quantile dependent prior distributions for the regression coefficients, assigning suitable quantile dependent priors over model space and achieving computational efficiency. The first of these challenges is studied in Chapter 2 in which a quantile dependent prior elicitation scheme is developed. In particular, an extension of the Zellners prior which allows for a conditional conjugate prior and quantile dependent prior on Bayesian quantile regression is proposed. The prior is generalised in Chapter 3 by introducing a ridge parameter to address important challenges that may arise in some applications, such as multicollinearity and overfitting problems. The proposed prior is also used in Chapter 4 for subset selection of the fixed and random coefficients in a linear mixedeffects QR model. In Chapter 5 we specify normal-exponential prior distributions for the regression coefficients which can provide adaptive shrinkage and represent an alternative model to the Bayesian Lasso quantile regression model. For the second challenge, we assign a quantile dependent prior over model space in Chapter 2. The prior is based on the percentage bend correlation which depends on the quantile level. This prior is novel and is used in Bayesian regression for the first time. For the third challenge of computational efficiency, Gibbs samplers are derived and setup to facilitate the computation of the proposed methods. In addition to the three major aforementioned challenges this thesis also addresses other important issues such as the regularisation in quantile regression and selecting both random and fixed effects in mixed quantile regression models.

Regression testing is essential to ensure software quality. Regression Test-case selection is another process wherein, the testers would like to ensure that test-cases which are obsolete due to the changes in the system should not be considered for further testing. This is the Regression Test-case Selection problem. Although existing research has addressed many related problems, most of the existing regression test-case selection techniques cater to procedural systems. Being academic, they lack the scalability and detail to cater to multi-tier applications. Such techniques can be employed for procedural systems, usually mathematical applications. Enterprise applications have become complex and distributed leading to component-based architectures. Thus, inter-process communication has become a very important activity of any such system. Messaging is the most widely employed intermodule interaction mechanism. Today's systems, being heavily internet dependent, are Web-Services based which utilize XML for messaging. We propose an RTS technique which is specifically targeted at enterprise applications.

M.Comm. / This study analyzes the effect of the subprime crisis on portfolio allocation from the perspective of dependence structure. Empirical evidence has proved that the multivariate normal distribution is inadequate to model portfolio asset return distribution - firstly because the empirical marginal distributions of asset returns are skewed and fat tailed; and secondly because it does not consider the possibility of extreme joint co-movement of asset returns (Fama and French, 1993; Richardson and Smith, 1993; Géczy, 1998; Longin and Solnik, 2001; Mashal and Zeevi, 2002). This study employs Archimedean copulas to capture both the dependence structure and the asymmetry of asset returns in the tails of the empirical distributions.

Die Auswahl des attraktivsten Portfolios von Investitionsprojekten zählt zu den kritischen Ma-nagementaufgaben. Angesichts mehrfacher Zielsetzungen und komplexer Projektabhängigkeiten bietet sich dazu ein zweistufiges Vorgehen an, das zunächst effiziente Portfolios identifiziert und den Entscheidungsträger anschließend bei der Suche in diesem Lösungsraum unterstützt. Bei einer großen Zahl an Vorschlägen können die möglichen Projektkombinationen aber nicht mehr in akzeptabler Zeit vollständig enumeriert werden. Adaptierte Meta-Heuristiken bieten hier einen Kompromiß zwischen dem Wunsch nach exakter Bestimmung aller Pareto-optimalen Investitionsprogramme und dem dazu nötigen Rechenaufwand. Dieser Beitrag beschreibt den entsprechenden Einsatz künstlicher Ameisen und diskutiert erste numerische Ergebnisse. / Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

<p>The concept of biological fitness is foundational for our understanding of both ecology and evolution. Fitness is often described vaguely as an organism’s contribution to the next generation. The reason this is vague is because researchers define and measure fitness differently across fields. I suggest that the myriad definitions and ways to measure fitness commonly employed have led to debates and, seemingly contradictory results. In order to investigate the use of the concept of fitness, I performed a literature review and asked, (1) How is biological fitness defined and used by researchers? (2) How is fitness actually measured by researchers? To address these questions, I surveyed 478 papers published between 2012 and 2016, that included the word ‘fitness’ in the title, and were in the Web of Science categories of ‘ecology’ and ‘evolutionary biology’. In my analysis of the journal articles fitness was only defined 33% of the time. Among studies that did explicitly define fitness, I categorized 18 different definitions, though only 7 were found in more than 5% of papers. I also found differences in how fitness was measured. I found 87 measurements that I grouped into 13 categories. In addition to my survey of the literature, I performed an experiment to explore the relationship between different measures of fitness. Vegetative biomass and reproductive biomass are often both used as metrics of fitness by plant ecologists. In this experiment I determined the relationship between two popular measures of plant fitness vegetative biomass and reproductive yield. I found that these two proxies for plant fitness, vegetative biomass and reproduction, were unimodally related, meaning: 1) intermediate sized plants have the greatest reproductive output, and; 2) for any unique amount of reproduction there is both a small and a large plant with identical reproductive output. Two things emerge from the literature review and the experiment: first, given the many definitions that exist, researchers should be clear about which one they are using. Second, one must be clear about the expected relationship between proxy measurements and fitness, as it may be complex, or non-existent.</p><p></p>