A geographic information system (GIS) was used to create landform measurements and maps for elevation, slope, aspect, landform index, relative phenologic change, and slope position for 3 topographic quadrangles in Virginia. A set of known observation points of the Northern dusky flying squirrel (Glaucomys sabrinus) was used to build 3 models to delineate sites with landform characteristics equivalent to those known points. All models were built using squirrel observation points from 2 topographic quadrangles. The first model, called "exclusionary", excluded those pixels with landform characteristics different from the known squirrel pixels based on histogram analyses. Logistic regression was used to create the other 2 models. Each model resulted in an image of pixels considered equivalent to the known squirrel pixels. Each model excluded approximately 65% of the Highland study area, but the exclusionary model excluded the fewest known squirrel pixels (12.62%). Both logistic regression models excluded approximately 10% more known squirrel pixels than the exclusionary approach. The models were tested in the area of a third quadrangle with points known to be occupied by squirrels. After the model was applied to the third topographic quadrangle, the exclusionary model excluded the least amount of full-area pixels (79.30%) and only 14.81% of the known squirrel pixels. The second logistic regression excluded 81.16 % of the full area and no known squirrel pixels. All models proved useful in quickly delineating pixels equivalent to areas where wildlife were known to occur. / Master of Science

Georeferenced field data were used to develop logistic regression models of the geographic distribution of 38 frequently common plant species throughout Rich County, Utah, to assist in the future correlation of Natural Resources Conservation Service Ecological Site Descriptions to soil map units. Field data were collected primarily during the summer of 2007, and augmented with previously existing data collected in 2001 and 2006. Several abiotic parameters and Landsat Thematic Mapper imagery were used to stratify the study area into sampling units prior to the 2007 field season. Models were initially evaluated using an independent dataset extracted from data collected by the Bureau of Land Management and by another research project conducted in Rich County by Utah State University. By using this independent dataset, model accuracy statistics widely varied across individual species, but the average model sensitivity (modeling a species as common where it was common in the independent dataset) was 0.626, and the average overall correct classification rate was 0.683. Because of concerns pertaining to the appropriateness of the independent dataset for evaluation, models were also evaluated using an internal cross-validation procedure. Model accuracy statistics computed by this procedure averaged 0.734 for sensitivity and 0.813 for overall correct classification rate. There was less variability in accuracy statistics across species using the internal cross-validation procedure. Despite concerns with the independent dataset, we wanted to determine if models would be improved, based on internal cross-validation accuracy statistics, by adding these data to the original training data. Results indicated that the original training data, collected with this modeling effort in mind, were better for choosing model parameters, but sometimes model coefficients were better when computed using the combined dataset.

Utilizing survey data collected in four communities in the State of Utah, this study examined the extent to which rural resident perceptions and attitudes toward local community circumstances influence their own expectations and attitudes subjectively toward future community development alternatives. Understanding perceptions of community and community development, as well as the patterns of localized community development, is crucial and needs to consider residents' opinions and attitudes toward unique rural economic, environmental, and social conditions in order to help preserve the unique characteristics of the way of life while continuing economic improvement and social betterment in rural areas. Three conceptual frameworks of development (economic, environmental, and social) are applied in this study to explore the relationship between local residents' general attitudes toward the current conditions in their community and their attitudes toward development alternatives. I examine how these three development frameworks guide rural scholars to understand whether the pattern of community development is consistent across the region or localized from community to community. Four different types of rural communities were selected in a Utah-wide community survey in the summer of 2008. These communities are facing four different change patterns: an increasing senior community, an energy-development community, a recreational community, and a constant community that has remained stable over the last five decades. Each type of community has its unique economy, lifestyle, culture, and environment, in which local residents have developed a way of life in response to these changes in social and economic structures. Research findings indicate that the local residents' self-perceptions of community economic situation are not significant indictors to support the arguments of the economic development framework. However, indexes of environmental and social development frameworks are found to have strong associations with locals' environmental and social development alternatives. Also, different types of rural community show different demands for community development strategies, implying that a single development framework would not be sufficient to explain the complex of local residents' perceptions and attitudes toward community development unless the researchers integrate other perspectives into the model.

abstract: Optimal design theory provides a general framework for the construction of experimental designs for categorical responses. For a binary response, where the possible result is one of two outcomes, the logistic regression model is widely used to relate a set of experimental factors with the probability of a positive (or negative) outcome. This research investigates and proposes alternative designs to alleviate the problem of separation in small-sample D-optimal designs for the logistic regression model. Separation causes the non-existence of maximum likelihood parameter estimates and presents a serious problem for model fitting purposes. First, it is shown that exact, multi-factor D-optimal designs for the logistic regression model can be susceptible to separation. Several logistic regression models are specified, and exact D-optimal designs of fixed sizes are constructed for each model. Sets of simulated response data are generated to estimate the probability of separation in each design. This study proves through simulation that small-sample D-optimal designs are prone to separation and that separation risk is dependent on the specified model. Additionally, it is demonstrated that exact designs of equal size constructed for the same models may have significantly different chances of encountering separation. The second portion of this research establishes an effective strategy for augmentation, where additional design runs are judiciously added to eliminate separation that has occurred in an initial design. A simulation study is used to demonstrate that augmenting runs in regions of maximum prediction variance (MPV), where the predicted probability of either response category is 50%, most reliably eliminates separation. However, it is also shown that MPV augmentation tends to yield augmented designs with lower D-efficiencies. The final portion of this research proposes a novel compound optimality criterion, DMP, that is used to construct locally optimal and robust compromise designs. A two-phase coordinate exchange algorithm is implemented to construct exact locally DMP-optimal designs. To address design dependence issues, a maximin strategy is proposed for designating a robust DMP-optimal design. A case study demonstrates that the maximin DMP-optimal design maintains comparable D-efficiencies to a corresponding Bayesian D-optimal design while offering significantly improved separation performance. / Dissertation/Thesis / Doctoral Dissertation Industrial Engineering 2019

The use of machine learning methods in credit risk modelling has been proven to yield good results in terms of increasing the accuracy of the risk score as- signed to customers. In this thesis, the aim is to examine the performance of the machine learning boosting algorithms XGBoost and CatBoost, with logis- tic regression as a benchmark model, in terms of assessing credit risk. These methods were applied to two different data sets where grid search was used for hyperparameter optimization of XGBoost and CatBoost. The evaluation metrics used to examine the classification accuracy of the methods were model accuracy, ROC curves, AUC and cross validation. According to our results, the machine learning boosting methods outperformed logistic regression on the test data for both data sets and CatBoost yield the highest results in terms of both accuracy and AUC.

In this thesis the relation between gender diversity and innovation in technology and manufacturing companies is explored. Data on firm-level are used from The Enterprise Surveys of the World Bank, which are designed as a panel data survey and comprise a collection of data on 146,000 firms in 143 countries, from years 2006 to 2019. Our focus group is technology and manufacturing firms, therefore, the final data used comprises of 8,839 firms in 47 countries for the year 2013 to investigate whether gender diversity is positively related to innovation of technology and manufacturing firms. Binary logistic regression analysis is used due to the nature of the available data measuring innovation output, which is the survey answer whether a firm introduced a new product/process or not. There are controversies in current research findings caused by the classification of incremental and radical innovation, therefore, this thesis takes an inclusive approach that accounts for total innovation (both incremental and radical). We also assess the total innovation in terms of new products and processes. Gender diversity is measured as total gender diversity of the permanent full-time employees and also on top management level. We also control for industry type, firm size, firm structure, firm’s export activity, R&D investment and employee training. The results show that there are currently low levels of gender inclusion on various firm levels globally. The regression analysis shows that only female presence on top management level made a unique statistically significant contribution to the model, and not total gender diversity on employee level. Regarding the control variables, only firm size, having invested in R&D, and offering employee training made a unique statistically significant contribution to the model. Conclusively, we found that gender diversity on top management level is positively related to innovation performance of technology and manufacturing firms, but not on employee level. However, due to the nature of panel data surveys when it is not possible to lag the cause with respect to the effect, a ‟cause-effect” relationship cannot be deduced with confidence. Nevertheless, our results are in line with the existing theory which indicates that gender diversity on leadership level may have a small but positive effect on achieving firm goals and innovative ideas-decisions-strategies. An explanation why we did not find a positive relationship on employee level can be the fact that during the innovation process the role of individuals and thus gender is invisible - hidden within processes, organizations, systems and there is lack of separating creativity from implementation; employee diversity might improve the creative process but impede the implementation. It is probably easier to assess the role of individuals on top management level and compare the effect of different leadership styles across companies. For external observers this assessment seems more complicated on employee level, thus the benefits of gender diversity even on employee level should not be underestimated. Therefore, more gender diversified quota and policies may need to be taken by decision makers with potentially positive impact both on society and economy. The relationship between innovation and gender diversity is a rather complex subject, affected by many internal and external firm contexts. By accounting for control variables including firm size, structure, export activity, R&D investment, employee training and industry type, some possible causal factors are eliminated. As prior research has already indicated, other factors that have not been addressed yet (and not covered by our framework either) are firm level structures, capabilities, innovation strategies, management style, team structure-functional diversity.

Tranquilizer misuse is a behavior associated with substance use disorder. As of now there is only one published article that includes a predictive model on misuse of subscription tranquilizers. The aim of this study is to predict ongoing tranquilizer misuse whilst comparing three different methods of classification; (1) regularized logistic regression, (2) adaptive boosting and (3) support vector machines. Data from the National Survey of Drug Use and Health (NSDUH) from 2019 is used to predict misuse among the individuals in the sample from 2020. The regularized logistic regression and the support vector machines models both yield an AUC of 0.88, which is slightly higher than the adaptive boosting model. However, the support vector machine model yields a higher level of sensitivity, meaning that it is better at detecting individuals who misuse. Although the difference in performance between the methods is relatively small and is most likely caused by the fact that different methods perform differently depending on the characteristics of the data.

京都大学 / 0048 / 新制・課程博士 / 博士(社会健康医学) / 甲第19638号 / 社医博第71号 / 新制||社医||9(附属図書館) / 32674 / 京都大学大学院医学研究科社会健康医学系専攻 / (主査)教授 中山 健夫, 教授 今中 雄一, 教授 松原 和夫 / 学位規則第4条第1項該当 / Doctor of Public Health / Kyoto University / DFAM

No description available.

No description available.