Age-Period-Cohort models offer a useful framework to study trends of time-specific phenomena in various areas. Yet the perfect linear relationship among age, period, and cohort induces a singular design matrix and brings about the identification issue of age, period, and cohort model due to the identity Cohort = Period -- Age. Over the last few decades, multiple methods have been proposed to cope with the identification issue, e.g., the intrinsic estimator (IE), which may be viewed as a limiting form of ridge regression. This study views the ridge estimator from a Bayesian perspective by introducing a prior distribution(s) for the ridge parameter(s). Data used in this study describe the incidence rate of cervical cancer among Ontario women from 1960 to 1994. Results indicate that a Bayesian ridge model with a common prior for the ridge parameter yields estimates of age, period, and cohort effects similar to those based on the intrinsic estimator and to those based on a ridge estimator. The performance of Bayesian models with distinctive priors for the ridge parameters of age, period, and cohort effects is affected more by the choice of prior distributions. In sum, a Bayesian ridge model is an alternative way to deal with the identification problem of age, period, and cohort model. Future studies should further investigate the influences of different prior choices on Bayesian ridge models. / text

Mirtingumo nuo įvairių ligų įvertinimas ir prognozavimas pagal atlikto tyrimo duomenis – dažnas statistinės analizės uždavinys medicinoje. Juose siekiama prognozuoti tikėtiną mirčių nuo tiriamos ligos skaičių, susirgimo tam tikra liga tikimybę ar išskirti rizikos grupes, įvertinant tyrimo metu surinktų stebimos populiacijos imties kintamųjų duomenis ir nustatant, kokia priklausomybę juos sieja. Pagrindiniai šio darbo tikslai: susipažinti su statistikos metodais, taikomais mirtingumo duomenų analizei; sudaryti statistinės analizės modelius turimiems mirtingumo duomenims; realizuoti sudarytus modelius programiškai, panaudojant SAS sistemą ir SAS makro programavimo galimybes. Panaudojus Puasono, logistinės ir Kokso regresin��s analizės metodus sudaryti mirtingumo nuo galvos smegenų insulto (GSI) prognozavimo modeliai. Sudaryti modeliai realizuoti programiškai, panaudojus SAS programavimo kalbą, SAS/IML posistemės galimybes ir SAS makro programavimo priemones. Sukurti regresinės analizės modeliai ir programines priemonės panaudotos Kauno medicinos universiteto Kardiologijos instituto 1980-2004 metais atliktų tyrimų metu surinktų Kauno miesto 25-64 m. amžiaus gyventojų mirtingumo nuo GSI duomenų analizei atlikti. / Estimation and forecasting of mortality from various diseases are very frequent data analysis tasks in medicine nowadays. In order to estimate expected number of deaths, probability to die from a disease or trends in mortality we should apply the most suitable statistical methods. Data analysis models were created using Poisson, logistic, Cox regression methods and realized in SAS macros. Created software also contains models for goodness of fit analysis, graphical visualization and prepares a report of data analysis in RTF (Rich Text Format) format. Analysis was made for mortality from stroke data among Kaunas population aged 25 to 64 during the period 1980-2004. The study contains the description of applying created data analysis models, SAS macros and received results.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-10-06T17:59:14Z No. of bitstreams: 1 christianemariameureralves.pdf: 1154792 bytes, checksum: 93b03559e6d95acbb4dbb3661f857e03 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-10-07T12:14:33Z (GMT) No. of bitstreams: 1 christianemariameureralves.pdf: 1154792 bytes, checksum: 93b03559e6d95acbb4dbb3661f857e03 (MD5) / Made available in DSpace on 2016-10-07T12:14:33Z (GMT). No. of bitstreams: 1 christianemariameureralves.pdf: 1154792 bytes, checksum: 93b03559e6d95acbb4dbb3661f857e03 (MD5) Previous issue date: 2009-02-13 / Introdução: O câncer de colo de útero, desde a década de 50, dispõe de um exame capaz de detectá-lo em fase incipiente e curável. A disponibilidade do teste de Papanicolaou parece ser a principal motivação para a queda de mortalidade por câncer de colo de útero em vários países ao redor do mundo. Buscou-se com este estudo avaliar o comportamento da mortalidade por câncer de colo de útero e útero porção não especificada, no período de 1980-2005, no Estado de Minas Gerais. Optou-se pela utilização de modelo de regressão linear e pela abordagem idadeperíodo-coorte. Material e Métodos: Foram coletados os dados de óbito e população disponíveis no DATASUS. Para avaliação da tendência de mortalidade por idade e período, utilizou-se o modelo de regressão linear; as taxas também foram log-transformadas para que se obtivesse o percentual de mudança da mortalidade por ano. A análise período-coorte foi feita através do método não paramétrico de Tarone e Chu. Resultados: Encontrou-se queda na mortalidade por câncer de colo de útero e útero porção não especificada para a análise idade e período. A redução foi principalmente relacionada com os casos de câncer de útero porção não especificada. Na análise idade-período-coorte houve redução menor que a esperada para as coortes de 1901-1908 e 1921-1928. Houve redução maior que a esperada para as coortes de 1913-1920, 1929-1932, 1937-1946, 1949-1956, 19631970 e 1969-1976. Encontrou-se ainda redução maior que a esperada para o período de 2000-2001. Conclusão: Foi evidenciada a redução da mortalidade por câncer de colo de útero e útero porção não especificada no Estado de Minas Gerais no período estudado. Os achados mostram influência das coortes de nascimento sobre a queda da mortalidade. / Introduction: Cervical cancer has had since the fifties, an exam capable of detecting it in its early and curable stage. The availability of the Papanicolaou smear test seems to be the principal reason for the fall in the mortality due to cervical cancer in many countries throughout the world. The aim of this study was to assess the trends of the mortality due to cervical cancer and uterus not otherwise specified (NOS) in the period from 1980-2005, in the state of Minas Gerais. We opted for the model of linear regression and the age-period-cohort approach. Material and Methods: Data related to death and population available at DATASUS were collected. To assess the tendency of mortality by age and period the approach of linear regression was used; the taxes were also log transformed in order to obtain the percentage of change in the mortality by year. The period-cohort analysis was carried out using Tarone & Chu’s non parametric method. Results: A reduction in the mortality due to cervical cancer and uterus not otherwise specified (NOS) for the age and period analyzed was found. The reduction was mainly related with the cases of cancer of uterus not otherwise specified (NOS). In the age-period-cohort analysis the reduction was less than expected for the cohorts from 1901-1908 and 1921-1928. There was a reduction bigger than expected for the cohorts from 1913-1920, 1929-1932, 1937-1946, 19491956, 1963-1970 and 1969-1976. It was also found a bigger reduction than expected for the period from 2000-2001. Conclusion: The reduction in the mortality due to cervical cancer and uterus not otherwise specified (NOS) in the state of Minas Gerais was unmistakable in the period studied. The findings show the influence of the birth cohorts over the decrease in mortality.

Ce mémoire analyse l’évolution de la prévalence de la surqualification en emploi au Québec à l’aide des données des recensements de 1971 à 2016. Les travaux portant sur l'évolution de la surqualification au Québec n’ont traité au plus que de la période allant de 1990 à 2012 et ont peu pris en considération certains facteurs de différenciation importants (niveau de diplôme, domaine d’études, âge, cohorte, genre, immigration). Ce mémoire cherche à combler ces lacunes pour mieux comprendre de quoi relève cette évolution. L’étude des dynamiques de la surqualification se révèle particulièrement riche dans la mesure où elle permet de mettre à l’épreuve un large spectre de théories concernant les rapports entre l’éducation et l’emploi. On analyse d'abord les évolutions respectives et conjointes de la structure des diplômes au sein de la population active et de la structure de la qualification des emplois. Alors qu’en 1971 le nombre d’emplois hautement qualifiés dépassait le nombre de travailleurs disposant d’un niveau de diplôme correspondant, ce rapport s’est inversé au fil du temps. Dans le contexte d’une hausse rapide des niveaux de scolarité et d’une hausse moindre de la qualification des emplois, le nombre de diplômés du postsecondaire a dépassé le nombre d’emplois qualifiés correspondants, si bien que la surqualification apparaît de plus en plus comme une situation d’emploi structurellement inévitable pour une partie des travailleurs. Au fil des dernières décennies, les taux de surqualification au sein de la population active ont augmenté, en partie sous l’effet mécanique de la hausse de la diplomation et donc du nombre de travailleurs susceptibles de connaître une situation de surqualification, mais aussi comme résultat d’une hausse des chances de surqualification à niveau de diplôme égal. Les dynamiques derrière cette hausse des chances de surqualification varient selon le niveau de diplôme. Dans l’ensemble, en ce qui a trait à l’accès à l’emploi qualifié, le diplôme devient de plus en plus nécessaire, et de moins en moins suffisant. Des modèles âge-période-cohorte sont utilisés pour estimer les effets respectifs des différentes dimensions de l’évolution dans le temps. Des effets de cohorte influençant les probabilités de surqualification des travailleurs sont observables. Les effets de l’âge prennent une forme en « U », les plus jeunes et les plus âgés étant les plus à risque de connaître une situation de surqualification. Les différences dans les évolutions de la surqualification en fonction du genre et du fait d’être né au Canada ou hors du Canada sont aussi examinées. / This thesis analyzes the evolution of the prevalence of overqualification in Quebec using census data from 1971 to 2016. Until now, studies on the evolution of overqualification in Quebec have covered at most the period from 1990 to 2012 and have paid little attention to certain important factors of differentiation (education level, field of study, age, cohort, gender, immigration). This thesis seeks to fill this gap. Studying the dynamics of overqualification is particularly valuable in that it allows us to test a wide variety of theories regarding the relationships between education and employment. I first analyze changes over time in the structure of educational qualifications in the workforce and in the structure of the skill levels of jobs, as well as in the overlap between these structures. Whereas in 1971 the number of high skilled jobs was higher than the number of workers holding corresponding degrees, this situation was reversed over time. Due to the rapid rise in educational attainment and the slower rise in the skill level of jobs, the number of postsecondary graduates has outpaced the number of corresponding jobs, meaning that overqualification is increasingly becoming structurally inevitable for some workers. Over the past decades, overqualification rates in the labour force have increased, in part simply as a result of the increase in educational attainment and therefore of the number of workers for whom it is possible to experience overqualification, but also as a result of an increase in the risks of overqualification among graduates with the same level of education. Overall, when it comes to access to high skilled jobs, educational credentials are becoming both more necessary and less sufficient. Age-period-cohort models are used to estimate the respective effects of these different dimensions of change over time. Cohort effects on the probability of overqualification are observed. Age effects take a "U" shape, the youngest and oldest workers showing the highest risks of overqualification. Differences by gender and place of birth (Canada or outside Canada) are also examined.

Viele Studien zeigen, dass die Beteiligung an politischen Protesten in mittel- und osteuropäischen Ländern geringer ausfällt als in Westeuropa. Das Ausmaß und die Ursachen dieser Ost-West-Partizipationslücke werden jedoch immer noch debattiert. Diese Dissertation untersucht die Ursachen dieses europäischen Protestgefälles. Inspiriert von den Theorien politischer Sozialisation wird untersucht, inwiefern ein frühes Erleben von (1) Repression und (2) Mobilisierung während der Transition zur Demokratie das Protestverhalten verschiedener Generationen in Mittel- und Osteuropa geprägt hat. Hierfür werden mehrebenen Alters-Perioden-Kohorten-Modelle mit wiederholten länderübergreifenden Umfragedaten genutzt. Studie 1 zeigt, dass ein frühes Erleben von Repression einen nachhaltigen Effekt auf die Teilnahme an Demonstrationen hat, nicht aber auf Petitionen und Boykotte. Darüber hinaus beeinflusst die Art der erlebten Repression die Richtung des Effekts: Personen, deren Bürgerrechte während ihrer Jugend eingeschränkt wurden, scheinen in ihrem späteren Leben häufiger an Demonstrationen teilzunehmen. Das Gegenteil ist der Fall für Personen, die Verletzungen persönlicher Integrität erlebt haben. Studie 2 zeigt, dass das Erleben der Mobilisierung während der Transition zur Demokratie diese Ost-West-Protestlücke nicht moderiert. Studie 3, eine Analyse des Protestverhaltens von Ostdeutschen, bestätigt, dass die Erfahrung der bottom-up Transition die mit gewaltsamer Repression verbundene Demobilisierung nicht kompensiert. Durch diese neu gewonnen Erkenntnisse zum Verhältnis von Regimewechsel und Zivilgesellschaft, verbindet und bereichert diese Dissertation die Forschungsfelder zu politischem Verhalten, sozialen Bewegungen und Demokratisierung. / Many studies have shown that protest participation is lower in Central and Eastern Europe than in Western Europe. Yet, the extent of and causes underlying the East-West participation gap are still debated in the literature. This thesis sheds new light on the sources of the European protest divide. Inspired by political socialization theories, it examines how early exposure to (1) repression and (2) mobilization during the transition to democracy has shaped the protest behavior of different generations in postcommunist democracies. This projects applies multilevel age-period-cohort models on data from repeated cross-national surveys to measure the effects of these types of exposure. Study 1 reveals that early exposure to repression has a lasting effect on demonstration attendance but not on participation in petitions and boycotts. Furthermore, the direction of this effect depends on the type of repression experienced by citizens: early exposure to civil liberties restrictions increases citizens’ participation in demonstrations while exposure to personal integrity violations depresses their participation. Study 2 demonstrates that exposure to mobilization during the transition to democracy does not moderate the East-West protest gap. Study 3, an analysis of East Germans’ protest behavior, confirms that the experience of a bottom-up transition does not compensate for the demobilization associated with violent repression. By generating new insights into the relation between regime change and civil society, this project bridges and contributes to the fields of political behavior, social movements, and democratization.