Thesis (M.S.)--Georgia State University, 2006. / Title from title screen. Yu-Sheng Hsu, committee chair; Gengsheng (Jeff) Qin, Xu Zhang, committee members. Electronic text (116 p.) : digital, PDF file. Description based on contents viewed May 17, 2007. Includes bibliographical references (p. 61-67).

Modelos para dados binários são bastante utilizados em várias situações práticas. Transformações em Análise de Regressão podem ser aplicadas para linearizar ou simplificar o modelo e também para corrigir desvios de suposições. Neste trabalho, descrevemos o uso de transformações nos modelos de regressão logística para dados binários e apresentamos modelos envolvendo parâmetros adicionais de modo a obter um ajuste mais adequado. Posteriormente, analisamos o custo da estimação quando são adicionados parâmetros aos modelos e apresentamos os testes de hipóteses relativos aos parâmetros do modelo de regressão logística de Box-Cox. Finalizando, apresentamos alguns métodos de diagnóstico para avaliar a influência das observações nas estimativas dos parâmetros de transformação da covariável, com aplicação a um conjunto de dados reais. / Binary data models have a lot of utilities in many practical situations. In Regrssion Analisys, transformations can be applied to linearize or simplify the model and correct deviations of the suppositions. In this dissertation, we show the use of the transformations in logistic models to binary data models and models involving additional parameters to obtain more appropriate fits. We also present the cost of the estimation when parameters are added to models, hypothesis tests of the parameters in the Box-Cox logistic regression model and finally, diagnostics methods to evaluate the influence of the observations in the estimation of the transformation covariate parameters with their applications to a real data set.

Traffic from major hurricane evacuations is known to cause severe gridlocks on evacuation routes. Better prediction of the expected amount of evacuation traffic is needed to improve the decision-making process for the required evacuation routes and possible deployment of special traffic operations, such as contraflow. The objective of this dissertation is to develop prediction models to predict the number of daily trips and the evacuation distance during a hurricane evacuation. Two data sets from the surveys of the evacuees from Hurricanes Katrina and Ivan were used in the models' development. The data sets included detailed information on the evacuees, including their evacuation days, evacuation distance, distance to the hurricane location, and their associated socioeconomic characteristics, including gender, age, race, household size, rental status, income, and education level. Three prediction models were developed. The evacuation trip and rate models were developed using logistic regression. Together, they were used to predict the number of daily trips generated before hurricane landfall. These daily predictions allowed for more detailed planning over the traditional models, which predicted the total number of trips generated from an entire evacuation. A third model developed attempted to predict the evacuation distance using Geographically Weighted Regression (GWR), which was able to account for the spatial variations found among the different evacuation areas, in terms of impacts from the model predictors. All three models were developed using the survey data set from Hurricane Katrina and then evaluated using the survey data set from Hurricane Ivan. All of the models developed provided logical results. The logistic models showed that larger households with people under age six were more likely to evacuate than smaller households. The GWR-based evacuation distance model showed that the household with children under age six, income, and proximity of household to hurricane path, all had an impact on the evacuation distances. While the models were found to provide logical results, it was recognized that they were calibrated and evaluated with relatively limited survey data. The models can be refined with additional data from future hurricane surveys, including additional variables, such as the time of day of the evacuation.

SME exporters constitute a critical economic force that contributes significantly to national productivity and job creation in the Canadian economy. However, the academic literature suggests that female-owned SMEs are less likely to export. With lower export propensity, the potential of female-owned SMEs for organic growth, economic self-sufficiency and wealth creation could be comprised. This paper applies logistic regression to study factors that influence SME owners’ export propensity with particular reference to the moderating effect of gender in the context of the Ajzen and Fishbein ’s (2005) theory of Reasoned Action and Planned Behavior. We improve the methodology of prevailing research by redefining “gender” in a more appropriate way and by computing gender interaction effects more accurately. Based on this analysis, we found that, although male- and female-owned SMEs show different likelihoods of exporting, gender does not have a direct residual impact. Instead, systemic gender differences account for most differences in the export propensity between male-owned and female-owned SMEs. Specifically, female-owned SMEs may be systemically disadvantaged because their firms are smaller, more limited in management capacity with younger and less-experienced managers. The lack of resources and market knowledge become constraining factors for them with respect to becoming “export-ready”. Additionally, female SME owners show a higher perception of risk and financing difficulty (although they do not encounter higher rejection rates of financing applications). Their subjective perceptions of potential barriers may contribute to their reluctance to export.

The study aimed to developed a logistic model based on antibody proteins, vaccinations and demographic factors that predicts previous infection in Covid-19. The data set comprised of 2750 individuals from eldercare homes in Sweden, with four test dates executed between October of 2021 and August of 2022.  Exploratory data analysis revealed bimodal patterns in the antibodies against nucleocapsid protein within the non-infected group, raising suspicions of false negatives in the data. Due to the binary nature of the response and to be interpretable for further research, logistic regressions were used to model the relation between predictors and the logit of the response. Because of low performance scores and high probability for the presence of false negatives, K-means clustering algorithm was performed on the data. As a clustering variable, the logarithm of base 2 of the nucleocapsid protein was used, because of its theoretical relationship with previous infection in Covid-19.  Observations were reclassified using the clustering technique, and two new logistic models were fitted to the data. The final model contained polynomial terms to handle the non-linear relationship between the logit of the response and the predictors. We found a significant relationship between the logarithm of 2 of nucleocapsid protein and previous Covid-19 infection in the final model, with high prediction results. We reached an F1-score of 0.94, indicating a well-performing model.  Additionally, an algorithm was created to predict the days since infection, involving the change in nucleocapsid protein from one test date to the next, and a GAM model for fitting a smooth line to the data between nucleocapsid protein as response against the days since infection. Using this algorithm, we reached an absolute mean error between predicted results and actual days since infection of 23 days. This algorithm was later applied to observations reclassified in the clustering process.  In conclusion, the study successfully reclassified false negative observations with previous Covid-19 infection, and fitted a logistic model with high prediction score with F1-score of 0.94. Finally, an algorithm was created that estimated the days since infection with an absolute mean error of 23 days. / Syftet med studien var att utveckla en logistisk modell baserad på antikroppsproteiner, vaccinationer och demografiska faktorer som förutsäger tidigare infektion i Covid-19. Datamängden bestod av 2750 individer från äldreboenden i Sverige, med fyra testdatum utförda mellan oktober 2021 och augusti 2022.  Utforskande dataanalys visade på bimodala mönster i antikroppar mot nukleokapsidprotein inom den icke- infekterade gruppen, vilket gav upphov till misstankar om falskt negativa resultat i datamaterialet. På grund av svarets binära karaktär och för att vara tolkningsbara för vidare forskning användes logistiska regressioner för att modellera förhållandet mellan prediktorer och responsvariabeln. På grund av låga prediktionsresultat och hög sannolikhet av förekomsten av falskt negativa svar utfördes K-means-klusteralgoritmen på datat. Som klustervariabel användes logaritmen av bas 2 för nukleokapsidproteinet, på grund av dess teoretiska samband med tidigare infektion i Covid-19.  Observationerna omklassificerades med hjälp av klustertekniken, och två nya logistiska modeller anpassades till datat. Den slutliga modellen innehöll polynomiala termer för att hantera det icke-linjära förhållandet mellan responsens logit och prediktorerna. Vi fann ett signifikant samband mellan logaritmen av 2 av nuk- leokapsidprotein och tidigare Covid-19-infektion i den slutliga modellen, med ett högt prediktionsresultat. Vi nådde en F1-score på 0.94.  Dessutom skapades en algoritm som predicerade dagar sedan infektion med hjälp av förändringen i nukleokap- sidprotein från ett testdatum till nästa, och en GAM-modell för att anpassa ett glidande medelvärdeslinje till datat mellan nukleokapsidprotein som response mot dagarna sedan infektionen. Med hjälp av denna algoritm nåddes ett absolut medelfel på 23 dagar mellan prediktion och faktiskt tid sedan infektionen. Denna algoritm tillämpades senare på observationer som omklassificerats i klusterprocessen.  Sammanfattningsvis lyckades studien framgångsrikt omklassificera falskt negativa observationer med tidigare Covid-19-infektion och anpassade en logistisk modell med hög prediktionspoäng med en F1-score på 0.94. Slutligen skapades en algoritm som uppskattade dagarna sedan infektionen med ett absolut medelfel på 23 dagar.

Past studies concerning the effects of economic conditions on voter perceptions have tended to generalize their findings to the entire national electorate. Such generalizations fail to account for the different ideologies, lifestyles, and economic conditions that exist from state to state. In the current study, I compare the effects of subjective financial evaluations with the effects of objective economic indicators on voter perceptions and turnout in the state of Mississippi. The purpose is to determine the extent to which past findings on the national level hold up on the state level, with Mississippi as the subject of analysis. Using data from the Mississippi Poll and employing a logistic regression method, the findings show that Mississippian‟s perceptions of political figures are more strongly influenced by subjective financial evaluations. Voter turnout, on the other hand, was more strongly influenced by objective economic indicators than personal financial satisfaction.

No description available.

Colleges and universities are under mounting pressure to meet enrollment goals in the face of declining college attendance. Insight into student-level probability of enrollment, as well as the identification of features relevant in student enrollment decisions, would assist in the allocation of marketing and recruitment resources and the development of future yield programs. A logistic regression model was fit to predict which applicants will ultimately matriculate (enroll) at Concordia College. Demographic, geodemographic and behavioral features were used to build a logistic regression model to assign probability of enrollment to each applicant. Behaviors indicating interest (campus visits, submitting a deposit) and residing in a zip code with high alumni density were found to be strong predictors of matriculation. The model was fit to minimize false negative rate, which was limited to 18.1 percent, compared to 50-60 percent reported by comparable studies. Overall, the model was 80.13 percent accurate.

Predictive models have established a niche in the field of archaeology. Valued as tools in predicting potential archaeological sites, their use has increased with development of faster and more affordable computer technology. Predictive models highlight areas within a landscape where archaeological sites have a high probability of occurrence. Therefore, time and resources normally expended on archaeological exploration can then be more efficiently allocated to specified locations within a study area. In addition to the resulting predictive surface, these models also identify significant variables for site selection by prehistoric or historic groups. Relationships with the environment, whether natural or social, are extremely pertinent to strengthening the resource base. In turn, this information can be utilized to better interpret and protect valuable cultural resources. A predictive probability model was generated to locate Union Civil War fortifications around Petersburg, Virginia. This study illustrated the ease with which such analysis can be accomplished with the integrated use of a Geographic Information System with statistical analysis. Stepwise logistic regression proved effective in selecting significant independent variables to predict probabilities of fortifications within the study area, but faired poorly when applied to areas withheld from the initial building stage of the model. Variation of battle tactics between these two separate areas proved great enough to have a detrimental effect the model's effectiveness. / Master of Science

No description available.