Determinants of Regional Differentiation of Mortality in the Czech Republic Abstract There are considerable differences in socioeconomic and sociodemographic factors influencing mortality on the individual and also on the aggregate levels. These differences were observed and explained in many countries of the world. The objective of this thesis is to find and evaluate the most significant external factors which influence actual regional differentiation of mortality in districts of the Czech Republic. The objective was achieved by means of the demographic and statistical analysis methods. First, there was a comparison of mortality rates calculated for each of the socioeconomic clusters. It was found out that higher mortality rates appeared among men and women living in the districts with unfavourable external conditions. Using the Poisson log-linear model, the most important factors influencing differences in mortality rates in districts of the Czech Republic were identified. These factors are: share of the unemployed, share of the divorced, share of university-educated people and the number of physicans in hospitals per 1000 inhabitants. Share of the unemployed seems to be the most significant factor. There is a correlation between this factor and the total mortality rate as well as the leading causes of...

No description available.

No description available.

Safety is a continually significant issue in the construction industry. The Occupation Safety and Health Administration as well as individual construction companies are constantly working on verifying that their selected safety plans have a positive effect on reduction of workplace injuries. Worker safety is a large concern for both the workers and employers in construction and the government also attempts to impose effective regulations concerning minimum safety requirements. There are many different methods for creating and implementing a safety plan, most notably the Construction Industry Institute's (CII) Zero Accidents Techniques (ZAT). This study will attempt to identify a relationship between the level of ZAT implementation and safety performance on industrial construction projects. This research also proposes that focusing efforts on certain ZAT elements over others will show different safety performance results. There are three findings in this study that can be used to assist safety professionals in designing efficient construction safety plans. The first is a significant log-log relationship that is identified between the DEA efficiency scores and Recordable Incident Rate (RIR). There is also a significant difference in safety performance found between the Light Industrial and Heavy Industrial sectors. Lastly, regression is used to show that the pre-construction and worker selection ZAT components can predict a better safety performance. / Master of Science

Motor vehicle crashes are one of the leading causes of death in the United States. Traffic safety research targets at understanding the cause of crash, preventing the crash, and mitigating crash severity. This dissertation focuses on the driver-related traffic safety issues, in particular, on developing and implementing contemporary statistical modeling techniques on driving risk research on Naturalistic Driving Study data. The dissertation includes 5 chapters. In Chapter 1, I introduced the backgrounds of traffic safety research and naturalistic driving study. In Chapter 2, the state-of-practice statistical methods were implemented on individual driver risk assessment using NDS data. The study showed that critical-incident events and driver demographic characteristics can serve as good predictors for identifying risky drivers. In Chapter 3, I developed and evaluated a novel Bayesian random exposure method for Poisson regression models to account for situations where the exposure information needs to be estimated. Simulation studies and real data analysis on Cellphone Pilot Analysis study data showed that, random exposure models have significantly better model fitting performances and higher parameter coverage probabilities as compared to traditional fixed exposure models. The advantage is more apparent when the values of Poisson regression coefficients are large. In Chapter 4, I performed comprehensive simulation-based performance analyses to investigate the type-I error, power and coverage probabilities on summary effect size in classical meta-analysis models. The results shed some light for reference on the prospective and retrospective performance analysis in meta-analysis research. In Chapter 5, I implemented classical- and Bayesian-approach multi-group hierarchical models on 100-Car data. Simulation-based retrospective performance analyses were used to investigate the powers and parameter coverage probabilities among different hierarchical models. The results showed that under fixed-effects model context, complex secondary tasks are associated with higher driving risk. / Ph. D.

Malaria is one of the leading causes of mortality and morbidity globally. Land use/land cover (LULC) change have been found to affect the transmission and distribution of malaria in other regions, but no study has attempted to examine such relationships in Nepal. Therefore, this study was conducted in Nepal to assess LULC change between 2000 and 2010, to study the spatial and temporal trend of malaria incidence rate (MIR) between 1999 and 2015, and to understand the relationship between LULC and malaria. The land cover types used for this study are forest, water bodies, agriculture, grassland, shrubland, barren areas, built-up areas and paddy areas. Change detection techniques were used to study LULC change. The temporal trend of MIR in 58 districts, and the relationship between MIR and LULC were evaluated using Poisson and negative binomial regression. Forest, water bodies, snow cover, and built-up area increased in Nepal by 28.5%, 2.96%, 55.12% and 21.19% respectively while the rest of the LULC variables decreased. MIR decreased significantly in 21 districts; however, four districts namely Pyuthan, Kaski, Rupandehi and Siraha had a significantly increasing trend of MIR. During 2001, 2002, and 2003, MIR was positively related to water bodies and paddy areas. Similarly, MIR of 2010 was negatively related to grassland. However, there was no relationship between LULC and MIR in 2000, 2011, 2012 and 2013. It may be because MIR is decreasing significantly in the country and thus the influence of LULC change is also decreasing. / MS

1 Abstract: The present work investigates techniques of insurence ratemaking accor- ding to the claims counts of policyholders on the basis of information contained in policies. At the beginning, we provide a closer examination of the theory of genera- lized linear models, which have wide range of applications in the field of actuarial modeling. The second chapter presents the basic Poisson regression model as well as some particular verification methods. Specifically, deviance and Wald test could be found here and furthermore also important results for residuals. The third chapter contains information on alternative approaches to modeling the claim frequencies and at the end the GEE method, that can be applied in case of panel data, is de- scribed. The numerical study based on real insurace data in last part of this diploma thesis illustrate's previously described techniques which were obtained with the help of statistical software SAS.

O conhecimento antecipado das condições meteorológicas poderá ajudar a sociedade a evitar prejuízos e desperdícios de recursos humanos e materiais. Portanto, o objetivo deste estudo foi obter a partir de uma análise estatística um modelo capaz de predizer internações a partir dos dados de poluição do ar e índices biometeorológicos. Para isso, foram utilizados dados diários de 1997 a 2000, referentes à cidade de São Paulo. Os dados de internações por doenças respiratórias foram divididos em três categorias: AVAS (Afecções Vias Aéreas Superiores), AVAI (Afecções das Vias Aéreas Inferiores) e IP (Influenza e Pneumonia), estes dados foram obtidos junto ao Ministério da Saúde. Os dados referentes à poluição foram obtidos junto à CETESB (Companhia de Tecnologia de Saneamento Ambiental) e os dados meteorológicos foram obtidos da estação meteorológica do Parque Estadual das Fontes do Ipiranga. Os índices de conforto térmico foram descritos com base em variáveis meteorológicas. Através de uma metodologia estatística de Regressão de Poisson e Análise de Componentes Principais (ACP), encontraram-se modelos estatísticos capazes de prever em média internações por doenças respiratórias. Esses modelos foram nomeados MBCS (Modelo Brasileiro de Clima e Saúde). A ACP foi utilizada a fim de corroborar a modelagem de regressão. Os resultados encontrados mostraram associação entre AVAS e SO2, CO (ambos sem defasagem) e com o índice biometeorológico TEv4 (com defasagem de 4 dias). Os resultados chamam atenção para o SO2 que, mesmo muito abaixo do padrão de qualidade do ar recomendado, ainda provoca acréscimos nas internações. Para as AVAI, os resultados mostram associações entre os poluentes MP10, O3 (ambos sem defasagem) e TEv4 (com 3 dias defasamento). Com relação à IP, as variáveis que se mostraram relacionadas foram MP10 (sem defasagem) e TEv4 (com 3 dias defasagem). Para verificar o skill do modelo, utilizou-se o ano de 2001. Os modelos apresentaram erro médio de 15% para AVAS, 30% para AVAI e 44% para IP com relação à previsão das internações. No que diz respeito a ACP, esta concorda com o que foi encontrado na modelagem de Poisson. Porém para AVAI e IP, os escores dos poluentes e dos índices deverão ser usados separadamente. Estes resultados mostram que o MBCS poderá ser utilizado para previsão de internação, contribuindo para políticas públicas e os meios de comunicação, ajudando nas tomadas de decisões e evitando desperdícios econômicos e humanos. / The meteorological condition knowledge can provide society prejudice prevention regarding human and material resources. Therefore, the aim of this study the statistical modeling in order to prevent internment of morbidity based on air pollution and meteorological variability. The whole 1997 to 2000 at the city of São Paulo. The morbidity data was divided in to three categories: AVAS (upper respiratory airway diseases), AVAI (lower respiratory airway diseases) and IP (Influenza and Pneumonia). These data were obtained from Brazilian Heath Ministry. Air pollution data were obtained from CETESB (Environmental agency) and meteorological data from Parque Estadual das Fontes do Ipiranga. Thermal comfort indexes were also used based on meteorological variables. Poisson regression models as well as Principal Component (PC) models were used in order to evaluate the data through statistical methodology. These models were nominated MBCS (Brazilian Climate and Health Model). Scores from PC statistical analysis were also used in order to compare to multiple regression models. As the first results, AVAS modeling presents association with SO2, CO (both without time lag) and the TEv4- a biometeorological index (with 4 days time lag). SO2 presents interesting result due to the fact that it is below the recommended standard, but it still causes AVAS morbidity. Concerning AVAI results, the variables which explain the morbidity were the pollutants MP10, O3 (both without lag) and TEv4 (with 3 days lag). Regarding to the skill of the models, AVAS model presents a 15% average error; AVAI model, 30% and IP model, 44%, during year of 2001. PC analysis corroborated the Poisson models. Regarding PC more weight for AVAS was pollutants. Already AVAI and IP more weight was biometeorological indexes and meteorological variables. The risk results used scores was similar to the MMRP. However for AVAI and IP, the scores of the pollutants and scores of the indexes should be used individually. These results indicate these models can be used as a forecasting internment program, contributing on the public and media decisions, avoiding economical and human unnecessary wastes.

In this dissertation, several essays in the field of bio-security are presented. The estimation of the probability of an FMD outbreak by type and location of premises is important for decision making. In Essay I, we estimate and predict the probability/risk of an FMD outbreak spreading to the various premises in the study area. We first used a Poisson regression model with adjustment dispersion associated with random simulation results from the AusSpead model to estimate the parameters of the model. Our estimation and prediction show that large cattle loss could be concentrated in three counties-Deaf Smith, Parmer, and Castro. These results are based on approximately 70% of the feedlots with over 10,000 cattle located in the three counties previously mentioned. In Essay II, our objective is to determine the best mitigation strategies in minimizing animal loss based on AusSpead simulation model. We tested 15 mitigation strategies by using multiple comparison. The results show that the best mitigation strategies for all four scenarios are regular surveillance, slaughter of the infected animals, and early detection. We then used the Mixed Integer Programming to estimate costs of disposing of animal carcasses and transportation. Results show that the unit disposal cost will vary with carcass scale and the unit transportation cost also varies with the distribution of the infected premises and disposal locations. FMD seems to have varying impacts on equity markets. In Essay III, we studied returns at three different levels of the stock market. We determined results in a structural break, and then estimated the impact of the announcement of confirmed cases of FMD disease on the volatility of stock market returns by using a GARCH-Mean model. Our results show that the structure break occurs on the day with the largest number of confirmed cases for meat product firms rather than the day of the first confirmed case. We found that the conditional volatilities over the FMD period are higher than those over the sample period. The announcement of confirmed cases had the largest marginal impact on meat products. Investors may always consider maintaining a portfolio consisting of index funds or hedge funds.

This thesis consists of three essays that address various aspects of the economics of maternal health care. The first two essays examine the determinants of utilization of maternal health care services in low-income countries, while the third essay examines the determinants of utilization of prenatal ultrasonography in Canada. The first essay examines the influence of prenatal attendance (as well as a wide array of observed individual-, household- and community-level characteristics) on a woman’s decision to give birth at a health facility or at home for thirty-two low-income countries (across Asia, Sub-Saharan Africa and Latin America). This empirical investigation employs the Demographic and Health Surveys (DHS) data and a two-level random intercept model. The results show that prenatal attendance has a substantial influence on the use of facility delivery in all three geographical regions. Women having four prenatal visits were 7.3 times more likely to deliver at a health facility than those with no prenatal care. The second essay addresses two related questions: what factors determine a woman’s decision to seek prenatal care; and are those the same factors that determine the frequency of care? This investigation also utilizes Demographic and Health Surveys (DHS) data for thirty-two low-income countries (across Asia, Sub-Saharan Africa and Latin America) and applies a two-part and multi-level model to that data. The results suggest that, though a wide range of factors influence both decisions, that influence varies in magnitude across the two decisions, as well as across the three geographical regions. The third essay examines the influence of various socioeconomic and demographic factors on the frequency of prenatal ultrasounds in Canada, while controlling for maternal risk profiles. This investigation utilizes data from the Maternity Experience Survey (MES) of the Canadian Perinatal Surveillance System and employs a count data regression model (the Poisson distribution) to estimate the effect of various factors on the number of prenatal ultrasounds. The results of this investigation suggest that, even after controlling for maternal risk factors, the type of health-care provider, province of prenatal care, and timings of first ultrasound are the strongest predictors of number of ultrasounds.