This thesis has investigated two-stage regularized logistic regressions applied on the credit scoring problem. Credit scoring refers to the practice of estimating the probability that a customer will default if given credit. The data was supplied by Klarna AB, and contains a larger number of observations than many other research papers on credit scoring. In this thesis, a two-stage regression refers to two staged regressions were the some kind of information from the first regression is used in the second regression to improve the overall performance. In the best performing models, the first stage was trained on alternative labels, payment status at earlier dates than the conventional. The predictions were then used as input to, or to segment, the second stage. This gave a gini increase of approximately 0.01. Using conventional scorecutoffs or distance to a decision boundary to segment the population did not improve performance. / Denna uppsats har undersökt tvåstegs regulariserade logistiska regressioner för att estimera credit score hos konsumenter. Credit score är ett mått på kreditvärdighet och mäter sannolikheten att en person inte betalar tillbaka sin kredit. Data kommer från Klarna AB och innehåller fler observationer än mycket annan forskning om kreditvärdighet. Med tvåstegsregressioner menas i denna uppsats en regressionsmodell bestående av två steg där information från det första steget används i det andra steget för att förbättra den totala prestandan. De bäst presterande modellerna använder i det första steget en alternativ förklaringsvariabel, betalningsstatus vid en tidigare tidpunkt än den konventionella, för att segmentera eller som variabel i det andra steget. Detta gav en giniökning på approximativt 0,01. Användandet av enklare segmenteringsmetoder så som score-gränser eller avstånd till en beslutsgräns visade sig inte förbättra prestandan.

<p>This dissertation investigates whether dual credit students' academic performance in subsequent university courses is comparable to that of non-dual-credit students. The study uses data from a Midwest regional public university over a ten-year period and employs propensity score matching and proportional odds ordinal logistic regression to create balanced comparison groups and analyze the results. The findings indicate that students who completed the prerequisite course as dual credit have similar grades in subsequent university courses as those who completed the prerequisite course as a university student. The study also identifies significant predictors of academic performance in subsequent university courses, such as sex, historically underserved groups status, high school GPA, and course subject, regardless of dual credit status. However, first-generation status, SAT scores, and the time between courses are not statistically significant predictors. These results suggest that dual credit students are as prepared for subsequent university courses as non-dual-credit students. Nevertheless, academic outcomes differ based on certain factors, which should be considered when designing student success initiatives and allocating resources.</p>

Background: Studies of gingival bleeding and the effects of antioxidants on extracellular matrix and immunologic and inflammatory responses provide a rationale for hypothesizing that antioxidants reduce the risk for gingival bleeding.Methods: This study evaluated the role of antioxidants as contributing risk factors for gingival bleeding utilizing the Third National Health and Nutrition Examination Survey (NAHNES III). A sample of 18,825 adults (20 to ≥ 90 years of age), with dental measurement and assessment of serum levels of antioxidants were included in the study. Gingival bleeding was defined as those who had more than 30 percent of gingival bleeding in 28 sites examined. SPSS version 11.0 software and Epi-info 2000 were used to perform the statistical analysis.Results: Using multiple logistic regression in five separate antioxidants, the study showed an association between increased plasma levels of vitamin C (ascorbic acid) and decreased risk for gingival bleeding (OR= 0.33; 95% CI 0.15 to 0.72). An inverse relationship was also found between gingival bleeding and serum levels of beta carotene (OR=1.93; 95% CI 1.05 to 3.54). However, negative association was found between gingival bleeding and vitamin A (OR=2.60; 95% CI 1.04 to 6.50). No statistically significant association was observed between gingival bleeding and serum levels in vitamin E (alpha tocopherol) and selenium.Conclusion: Antioxidants, vitamin C, vitamin A, and beta carotene, were significant risk factors for gingival bleeding. This should be emphasized for improving the oral health of the U.S. adult population.

This research was geared towards the investigation of management strategies (factors) that affect the performance ofMSMEs in Kenya. Many developed countries record a time in history when entrepreneurial activities led to revival of economical growth after decline. This implies MSMEs is a very vital sector especially for a developing country like Kenya. MSMEs stagnate and their performance is uncertain according to writers such as Namusonge, Management inadequacies have been suggested in several studies. The objectives of this research was to, 1. To identifY the critical management factors affecting the performance of MSMEs in Kenya; ii. To establish the process through which managerial factors affect the performance of a MSMEs in Kenya ; m. To determine the integrative effect of various management factors in the MSMES in Kenya; IV. To establish the effect of demographics and management factors on performance, v. To establish effects of external environment on internal management factors A conceptual model was formulated from the literature review showing relationships of the management strategies and the environment they operate in. These relationships became the basis for the hypotheses which were later tested. In chapter 4, a mini research (pilot study) was conducted in May 2007,whose main aim was to test the reliability and validity of the research instruments. The 36 questionnaires returned were analysed through descriptive method. Results obtained indicated the instruments were reliable and the results valid. A few corrections suggested were made. The major correction was addition of question 35 to collect financial information. The data collection was done between mid August and mid October 2007.In chapter 5, the researcher analysesd the results of the survey after receiving 180 questionnaires. Time was a constraint. In chapter 6, the hypotheses and conceptual model were analysed and the results obtained suggested that, most strategies did not affect the profitability separately but severally. The integrated effect of the management strategies and the associated factors had a higher impact on performance of the MSMES than any individual strategies. In chapter 7, the conclusions, summaries and Recommendations are given. / Business Management / D. Com. (Business Management and Policy)

近幾年來，國人自殺死亡率不斷提高，且自殺死亡從1997年起已連續多年列於國人十大死亡原因之一，所以自殺防治工作刻不容緩。本研究採用自殺防治中心在桃園縣六家居家服務單位(龍祥、中國、仁愛、紅十字、家輔及寬福)所做之問卷調查資料，目的在於找出何種特性者，BSRS5 (The Five-Item Brief Symptom Rating Scale)分數及自殺意念分數可能較高。本研究屬於時序性追蹤資料，自民國96年5月份起，由居服人員針對受測對象進行訪談，大約每隔兩週收集一次，總共進行四次。 針對問卷進行基本敘述性統計、單項排名分析以及交叉分析後發現，在人口特質方面，男女性比例相當，年齡層主要皆在65~84歲，教育程度以不識字及國小為主；在BSRS5五題排名方面，以第一題「睡眠困難(難以入睡或早醒)」的平均分數最高，第四題「覺得比不上別人」平均分數最低；由交叉分析的結果發現身體狀況為一個重要的變數，身體狀況差的人BSRS5總分6分以上或自殺意念2分以上明顯較多。 對資料配適廣義估計方程式及Alternating Logistic Regressions的結果，發現在反應變數為BSRS5總分時，女性、身體狀況差及曾經看過精神科者BSRS5分數達到6分以上的可能性較高。若反應變數為自殺意念時，無論是利用廣義估計方程式或Alternating Logistic Regressions，從模型配適的結果發現只有BSRS5的效應顯著。不管利用BSRS5總分或是各題分開來看，BSRS5對自殺意念是一個相當有效的檢測工具，BSRS5分數愈高則自殺意念2分以上的機會也愈高。此外利用多層結構分析方法配適廣義估計方程式，針對BSRS5與受測次數間的關聯性分析，發現與配適傳統unstructured相關性矩陣的估計結果差異不大，但是可以減少許多參數估計，並且在電腦計算時間上是較快速的。 / In Taiwan, suicide has been among the top ten causes of death since 1997, and suicide prevention has thus attracted much attention since. Using the data provided by Taiwan Suicide Prevention Center (TSPC), this study is aimed to find out possible personal characteristics that might have some impacts on the BSRS5 (the Five-Item Brief Symptom Rating Scale) and suicide ideation scores The data come from a longitudinal study in which subjects from six elderly home service centers in Taoyuan County, Taiwan were visited four times between May and July, 2007, about two weeks between each visit. The total number of subjects is 1981. The proportions of male and female are nearly the same, the age range is from 65 to 84, and most of them have only an elementary school degree. Preliminary analyses indicate that among the five items in BSRS5, insomnia (the first item) is ranked the highest, and inferiority (the fourth item) is the lowest. In addition, health status is highly correlated to the BSRS5 and suicide ideation scores, the worse the health status, the higher the BSRS5 and suicide ideation scores. Fitting the data with Generalized Estimating Equation (GEE) and Alternating Logistic Regressions models with respect to the BSRS5 score, we further find that female, those who have bad health status, and those who have ever consulted a psychiatrist have higher probability that the BSRS5 score is greater than 6. As far as the suicide ideation score is concerned, the BSRS5 score is the only covariate that is statistically significant, an indication that BSRS5 is a useful tool for screening subjects at risk of committing suicide. While the conclusions stay the same whether the data are analyzed through GEE with commonly used unstructured correlation structure or newly developed multiblock and multilayer correlation structure, the latter approach reduces the computer time significantly.

This research was geared towards the investigation of management strategies (factors) that affect the performance ofMSMEs in Kenya. Many developed countries record a time in history when entrepreneurial activities led to revival of economical growth after decline. This implies MSMEs is a very vital sector especially for a developing country like Kenya. MSMEs stagnate and their performance is uncertain according to writers such as Namusonge, Management inadequacies have been suggested in several studies. The objectives of this research was to, 1. To identifY the critical management factors affecting the performance of MSMEs in Kenya; ii. To establish the process through which managerial factors affect the performance of a MSMEs in Kenya ; m. To determine the integrative effect of various management factors in the MSMES in Kenya; IV. To establish the effect of demographics and management factors on performance, v. To establish effects of external environment on internal management factors A conceptual model was formulated from the literature review showing relationships of the management strategies and the environment they operate in. These relationships became the basis for the hypotheses which were later tested. In chapter 4, a mini research (pilot study) was conducted in May 2007,whose main aim was to test the reliability and validity of the research instruments. The 36 questionnaires returned were analysed through descriptive method. Results obtained indicated the instruments were reliable and the results valid. A few corrections suggested were made. The major correction was addition of question 35 to collect financial information. The data collection was done between mid August and mid October 2007.In chapter 5, the researcher analysesd the results of the survey after receiving 180 questionnaires. Time was a constraint. In chapter 6, the hypotheses and conceptual model were analysed and the results obtained suggested that, most strategies did not affect the profitability separately but severally. The integrated effect of the management strategies and the associated factors had a higher impact on performance of the MSMES than any individual strategies. In chapter 7, the conclusions, summaries and Recommendations are given. / Business Management / D. Com. (Business Management and Policy)

Text in English / An increasingly recognised prevention approach for substance use entails reduction in risk factors and enhancement of promotive or protective factors in individuals and the environment surrounding them during their growth and development. However, in order to enhance the effectiveness of this approach, continuous study of risk aspects targeting different cultures, social groups and mixture of society has been recommended. This study evaluated the impact of potential risk and protective factors associated with family management and relations on adolescent substance abuse in South Africa. Exploratory analysis and cumulative odds ordinal logistic regression modelling was performed on the data while controlling for demographic and socio-economic characteristics on adolescent substance use. The most intensely used substances were tobacco, cannabis, cocaine, heroin and alcohol in decreasing order of use intensity. The specific protective or risk impact of family management or relations factors varied from substance to substance. Risk factors associated with demographic and socio-economic factors included being male, younger age, being in lower education grades, coloured ethnicity, adolescents from divorced parents and unemployed or fully employed mothers. Significant family relations risk and protective factors against substance use were classified as either family functioning and conflict or family bonding and support. Several family management factors, categorised as parental monitoring, discipline, behavioural control and rewards, demonstrated either risk or protective effect on adolescent substance use. Some factors had either interactive risk or protective impact on substance use or lost significance when analysed jointly with other factors such as controlled variables. Interaction amongst risk or protective factors as well as the type of substance should be considered when further considering interventions based on these risk or protective factors. Studies in other geographical regions, institutions and with better gender balance are recommended to improve upon the representativeness of the results. Several other considerations to be made when formulating interventions, the shortcomings of this study and possible improvements as well as future studies are also suggested. / Psychology / M. A. (Psychology)

Text in English / An increasingly recognised prevention approach for substance use entails reduction in risk factors and enhancement of promotive or protective factors in individuals and the environment surrounding them during their growth and development. However, in order to enhance the effectiveness of this approach, continuous study of risk aspects targeting different cultures, social groups and mixture of society has been recommended. This study evaluated the impact of potential risk and protective factors associated with family management and relations on adolescent substance abuse in South Africa. Exploratory analysis and cumulative odds ordinal logistic regression modelling was performed on the data while controlling for demographic and socio-economic characteristics on adolescent substance use. The most intensely used substances were tobacco, cannabis, cocaine, heroin and alcohol in decreasing order of use intensity. The specific protective or risk impact of family management or relations factors varied from substance to substance. Risk factors associated with demographic and socio-economic factors included being male, younger age, being in lower education grades, coloured ethnicity, adolescents from divorced parents and unemployed or fully employed mothers. Significant family relations risk and protective factors against substance use were classified as either family functioning and conflict or family bonding and support. Several family management factors, categorised as parental monitoring, discipline, behavioural control and rewards, demonstrated either risk or protective effect on adolescent substance use. Some factors had either interactive risk or protective impact on substance use or lost significance when analysed jointly with other factors such as controlled variables. Interaction amongst risk or protective factors as well as the type of substance should be considered when further considering interventions based on these risk or protective factors. Studies in other geographical regions, institutions and with better gender balance are recommended to improve upon the representativeness of the results. Several other considerations to be made when formulating interventions, the shortcomings of this study and possible improvements as well as future studies are also suggested. / Psychology / M. A. (Psychology)