Browsing by Subject "database"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Advances in diapriid (Hymenoptera: diapriidae) systematics, with contributions to cybertaxonomy and the analysis of rRNA sequence data(2009-05-15) Yoder, Matthew JonDiapriids (Hymenoptera: Diapriidae) are small parasitic wasps. Though found throughout the world they are relatively unknown. A framework for advancing diapriid systematics is developed by introducing a new web-based application/database capable of storing a broad range of systematic data, and the first molecular phylogeny specifically focused at examining intrafamilial relationships. In addition to these efforts, a description of a new taxon is provided. Several advantages of digital description, including linking descriptions to an ontology of morphological terms, are highlighted. The functionality of the database is further illustrated in the production of a catalog of diapriid host associations. The hosts database currently holds over 450 association records, for over 500 named taxa (parasitoids and hosts), and over 180 references. Diapriids are found to be primarily endoparasitoids of Diptera emerging from the host pupa. Phylogenetic inference for a molecular dataset of 28S and 18S rRNA sequence data, derived from a diverse selection of diapriids, is accomplished with a new suite of tools developed for handling complex rRNA datasets. Several parsimony-based methodologies, including an alignment-free method of analyzing multiple sequences, are reviewed and applied using the new software tools. Diapriid phylogenetic relationships are shown to be broadly congruent with existing morphology-based classifications. Methods for analyzing typically excluded sequence data are shown to recover phylogenetic signal that would otherwise be lost and the alignment-free method performed remarkably well in this regard. Empirically, phylogenetic approaches that incorporate structural data were not notably different than those that did not.Item Applying Calibration to Improve Uncertainty Assessment(2013-08-02) Fondren, Mark EdwardUncertainty has a large effect on projects in the oil and gas industry, because most aspects of project evaluation rely on estimates. Industry routinely underestimates uncertainty, often significantly. The tendency to underestimate uncertainty is nearly universal. The cost associated with underestimating uncertainty, or overconfidence, can be substantial. Studies have shown that moderate overconfidence and optimism can result in expected portfolio disappointment of more than 30%. It has been shown that uncertainty can be assessed more reliably through look-backs and calibration, i.e., comparing actual results to probabilistic predictions over time. While many recognize the importance of look-backs, calibration is seldom practiced in industry. I believe a primary reason for this is lack of systematic processes and software for calibration. The primary development of my research is a database application that provides a way to track probabilistic estimates and their reliability over time. The Brier score and its components, mainly calibration, are used for evaluating reliability. The system is general in the types of estimates and forecasts that it can monitor, including production, reserves, time, costs, and even quarterly earnings. Forecasts may be assessed visually, using calibration charts, and quantitatively, using the Brier score. The calibration information can be used to modify probabilistic estimation and forecasting processes as needed to be more reliable. Historical data may be used to externally adjust future forecasts so they are better calibrated. Three experiments with historical data sets of predicted vs. actual quantities, e.g., drilling costs and reserves, are presented and demonstrate that external adjustment of probabilistic forecasts improve future estimates. Consistent application of this approach and database application over time should improve probabilistic forecasts, resulting in improved company and industry performance.