No description available.

No description available.

Bayesian Model Averaging (BMA) is a new area in modern applied statistics that provides data analysts with an efficient tool for discovering promising models and obtaining esti-mates of their posterior probabilities via Markov chain Monte Carlo (MCMC). These probabilities can be further used as weights for model averaged predictions and estimates of the parameters of interest. As a result, variance components due to model selection are estimated and accounted for, contrary to the practice of conventional data analysis (such as, for example, stepwise model selection). In addition, variable activation probabilities can be obtained for each variable of interest. This dissertation is aimed at connecting BMA and various ramifications of the multivari-ate technique called Reduced-Rank Regression (RRR). In particular, we are concerned with Canonical Correspondence Analysis (CCA) in ecological applications where the data are represented by a site by species abundance matrix with site-specific covariates. Our goal is to incorporate the multivariate techniques, such as Redundancy Analysis and Ca-nonical Correspondence Analysis into the general machinery of BMA, taking into account such complicating phenomena as outliers and clustering of observations within a single data-analysis strategy. Traditional implementations of model averaging are concerned with selection of variables. We extend the methodology of BMA to selection of subgroups of observations and im-plement several approaches to cluster and outlier analysis in the context of the multivari-ate regression model. The proposed algorithm of cluster analysis can accommodate re-strictions on the resulting partition of observations when some of them form sub-clusters that have to be preserved when larger clusters are formed. / Ph. D.

My overall objective in this dissertation was to develop more integrative perspectives of several aspects of academic motivation. Rarely have researchers and theorists examined a more comprehensive model of academic motivation that pools multiple constructs that interact in a complex and dynamic fashion (Kaplan, Katz, and Flum, 2012; Turner, Christensen, Kackar-Cam, Trucano, and Fulmer, 2014). The more common trend in motivation research and theory has been to identify and explain only a few motivation constructs and their linear relationships rather than examine complex relationships involving 'continuously emerging systems of dynamically interrelated components' (Kaplan et al., 2014, para. 4). In this dissertation, my co-author and I focused on a more integrative perspective of academic motivation by first reviewing varying characterizations of one motivation construct (Manuscript 1) and then empirically testing dynamic interactions among multiple motivation constructs using a person-centered methodological approach (Manuscript 2). Within the first manuscript (Chapter 2), a theoretical review paper, we summarized multiple perspectives of the need for autonomy and similar constructs in academic motivation, primarily autonomy in self-determination theory, autonomy supports, and choice. We provided an integrative review and extrapolated practical teaching implications. We concluded with recommendations for researchers and instructors, including a call for more integrated perspectives of academic motivation and autonomy that focus on complex and dynamic patterns in individuals' motivational beliefs. Within the second manuscript (Chapter 3), we empirically investigated students' motivation in science class as a complex, dynamic, and context-bound phenomenon that incorporates multiple motivation constructs. Following a person-centered approach, we completed cluster analyses of students' perceptions of 5 well-known motivation constructs (autonomy, utility value, expectancy, interest, and caring) in science class to determine whether or not the students grouped into meaningful 'motivation profiles.' 5 stable profiles emerged: (1) low motivation; (2) low value and high support; (3) somewhat high motivation; (4) somewhat high empowerment and values, and high support; and (5) high motivation. As this study serves as a proof of concept, we concluded by describing the 5 clusters. Together, these studies represent a focus on more integrative and person-centered approaches to studying and understanding academic motivation. / Ph. D.

This thesis considers a Flexible Manufacturing System (FMS) comprised of programmable machine tools which are capable of performing multiple operations and which are interconnected by computer-controlled automated material handling equipment. The specific problem addressed is a job-shop type of situation in which at least a given number of each type of job needs to be performed on a given set of machines. The flexibility in the system arises in the form that each job can be performed in a variety of ways with each possible manner of performing it, called an Alternate Routing Combination (ARC), being defined by specifying the number of operations needed and the associated machine sequence. The problem is to select a set of jobs and their associated ARCs to be performed, and schedule their operations on the machines so as to optimize various objectives such as minimizing makespan or maximizing machine utilization, or minimizing total flowtime. This problem is mathematically modeled, and heuristic algorithms are presented along with computational results for the case of minimizing the makespan. / M.S.

Network management is a problem that faces designers and operators of any type of network. Conventional methods of capacity planning or configuration management are difficult to apply directly to networks that dynamically allocate resources, such as Asynchronous Transfer Mode (ATM) networks and emerging Internet Protocol (IP) networks employing Differentiated Services (DiffServ). This work shows a method to generically classify traffic in an ATM network such that capacity planning may be possible. These methods are generally applicable to other networks that support dynamically allocated resources. In this research, Call Detail Records (CDRs) captured from a ¡§live¡¨ ATM network were successfully classified into three traffic categories. The traffic categories correspond to three different video speeds (1152 kbps, 768 kbps, and 384 kbps) in the network. Further statistical analysis was used to characterize these traffic categories and found them to fit deterministic distributions. The statistical analysis methods were also applied to several different network planning and management functions. Three specific potential applications related to network management were examined: capacity planning, traffic modeling, and configuration management. / Master of Science

Twentieth Century climatology was dominated by two great figures: Wladamir Köppen and C. Warren Thornthwaite. The first carefully developed climatic parameters to match the larger world vegetation communities. The second developed complex formulas of "Moisture Factors" that provided efficient understanding of how evapotranspiration influences plant growth and health, both for native and non-native communities. In the latter half of the Twentieth Century, the Sunset Magazine Corporation develop a purely empirical set of Garden Zones, first for California, then for the thirteen states of the West, now for the entire nation in the National Garden Maps. The Sunset Garden Zones are well recognized and respected in Western States for illustrating the several factors of climate that distinguish zones. But the Sunset Garden Zones have never before been digitized and examined statistically for validation of their demarcations. This thesis examines the digitized zones with reference to PRISM climate data. Variable coverages resembling those described by Sunset are extracted from the PRISM data. These variable coverages are collected for two buffered areas, one in northern California and one in southern California. The coverages are exported from ArcGIS 9.1 to SAS® where they are processed first through a Principal Component Analysis, and then the first five principal components are entered into a Ward's Hierarchical Cluster Analysis. The resulting clusters were translated back into ArcGIS as a raster coverage, where the clusters were climatic regions. This process is quite amenable for further examination of other regions of California / Master of Science

No / In visual search experiments, the subject looks for a target item in a display containing different distractor items. The reaction time (RT) to find the target is measured as a function of the number of distractors (set size). RT is either constant, or increases linearly, with set size. Here we suggest a two-stage model for search in which items are first selected and then recognized. The selection process is modeled by (a) grouping items into a hierarchical cluster tree, in which each cluster node contains a list of all the features of items in the cluster, called the object file, and (b) recursively searching the tree by comparing target features to the cluster object file to quickly determine whether the cluster could contain the target. This model is able to account for both constant and linear RT versus set size functions. In addition, it provides a simple and accurate account of conjunction searches (e.g., looking for a red N among red Os and green Ns), in particular the variation in search rate as the distractor ratio is varied.

Materials have typically been selected based on the familiarities and past experiences of a limited number of designers with a limited number of materials. Problems arise when the designer is unfamiliar with new or improved materials, or production processes more efficient and economical than past choices. Proper utilization of complete materials and processing information would require acquisition, understanding, and manipulation of huge amounts of data, including dependencies among variables and "what-if" situations. The problem of materials selection has been addressed with a variety of techniques, from simple broad-based heuristics as guidelines for selection, to elaborate expert system technologies for specific selection situations. However, most materials selection methodologies concentrate only on material properties, leaving other decision criteria with secondary importance. Factors such as component service environment, design features, and feasible manufacturing methods directly influence the material choice, but are seldom addressed in systematic materials selection procedures. This research addresses the problem of developing a systematic materials selection procedure that can be integrated with standard materials data bases. The three-phase methodology developed utilizes a group technology code and cluster analysis method for the selection. The first phase is of go/no go nature, and utilizes the possible service environment requirements of ferromagnetism and chemical corrosion resistance to eliminate materials from candidacy. In the second phase, a cluster analysis is performed on key design and manufacturing attributes captured in a group technology code for remaining materials. The final phase of the methodology is user-driven, in which further analysis of the output of the cluster analysis can be performed for more specific or subjective attributes. / Master of Science

No / Model interpretation is one of the key aspects of the model evaluation process. The explanation of the relationship between model variables and outputs is relatively easy for statistical models, such as linear regressions, thanks to the availability of model parameters and their statistical significance . For “black box” models, such as random forest, this information is hidden inside the model structure. This work presents an approach for computing feature contributions for random forest classification models. It allows for the determination of the influence of each variable on the model prediction for an individual instance. By analysing feature contributions for a training dataset, the most significant variables can be determined and their typical contribution towards predictions made for individual classes, i.e., class-specific feature contribution “patterns”, are discovered. These patterns represent a standard behaviour of the model and allow for an additional assessment of the model reliability for new data. Interpretation of feature contributions for two UCI benchmark datasets shows the potential of the proposed methodology. The robustness of results is demonstrated through an extensive analysis of feature contributions calculated for a large number of generated random forest models.