Πλοήγηση ανά Επιβλέπων "Kostaki, Anastasia"

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Α Β Γ Δ Ε Ζ Η Θ Ι Κ Λ Μ Ν Ξ Ο Π Ρ Σ Τ Υ Φ Χ Ψ Ω

Τώρα δείχνει 1 - 11 από 11

The changing shape of the fertility curve in modern populations
(06-06-2019) Karamanidis, Athanasios I.; Καραμανίδης, Αθανάσιος Η.; Athens University of Economics and Business, Department of Statistics; Papageorgiou, Ioulia; Psarakis, Stelios; Kostaki, Anastasia
The age specific fertility pattern has a typical shape common in all human populations through years. In order to describe this shape a number of parametric models have been proposed. In recent developed populations, the shape of the age specific fertility rates exhibits a deviation from the classical one.This thesis describes the evolution of the shape of the fertility curve (first birth order) in several developed populations as Greece, United Kingdom,Spain, Sweden, Bulgaria, France, Italy, Norway, U.S.A. and Ireland, using the models Peristera - Kostaki and Peristera - Kostaki Mixture. Furthermore,we discuss the possible causes that lead to the changing shape of the fertility curve in modern populations and we provide some concluding remarks.
The changing shape of the mortality curve in early adult ages
(02/28/2019) Giourmetaki, Eleni; Γιουρμετάκη, Ελένη; Athens University of Economics and Business, Department of Statistics; Papageorgiou, Ioulia; Psarakis, Stelios; Kostaki, Anastasia
This dissertation provides a presentation of the ways to examine the accident hump, that is the excess mortality in early adulthood. Some of these methods are related to the causes of death that lead to this excess mortality in these age groups. Moreover, some basic parametric models are presented which are used to study the accident hump. Finally, three parametric models are applied to empirical mortality data with the help of the statistical programming language, R.
Control Charts for Poisson Data
Athanasiou, Evangelia-Lydia; Athens University of Economics and Business, Department of Statistics; Kostaki, Anastasia
This thesis is a review of Control Charts for Data following the Poisson distribution. Firstly, the basic control chart theory and charts for all types of data are being introduced. Then, Shewhart control charts for data following the Poisson distribution are being presented: The modified u Control Chart as well as the c Chart for Nonconformities which is being compared to the Poisson Moving Average chart. The Zero Inflated Poisson Model which monitors processes with excessive 0 counts and the generalized zero inflated Poisson distribution, an extension of the ZIP model are introduced. Furthermore, several EWMA as well as CUSUM control charts (the Double EWMA, the PGWMA Chart, and finally FIR Poisson EWMA control charts) are examined. As far as CUSUM charts are concerned, the conditional and marginal performance of a Poisson CUSUM chart when the parameter is unknown and the CUSUM control chart based on the Poisson distribution compounded by a Geometric distribution are examined. Finally, Control Charts for Multivariate Poisson distributions are presented.
Description and exploration of age specific fertility pattern on the Netherlands, Austria, Bulgaria and Finland
(Athens University of Economics and Business, 11-2013) Androutsos, Athanasios; Kostaki, Anastasia
Thesis - Athens University of Economics and Business. Postgraduate, Department of Statistics
An impact of economic crisis on mortality in Greece
(Athens University of Economics and Business, 06-2015) Kavvadas, Stefanos N.; Kostaki, Anastasia
Thesis - Athens University of Economics and Business. Postgraduate, Department of Statistics
Methodology and applications of kernel techniques in mortality data
(Athens University of Economics and Business, 05-2001) Peristera, Paraskevi M.; Athens University of Economics and Business, Department of Statistics; Kostaki, Anastasia
Thesis - Athens University of Economics and Business. Postgraduate, Department of Statistics
Methods of expanding abridged life tables: evaluation and comparisons
(Athens University of Economics and Business, 05-2001) Panousis, Vangelis; Athens University of Economics and Business, Department of Statistics; Kostaki, Anastasia
Thesis - Athens University of Economics and Business. Postgraduate, Department of Statistics
Parametric models of mortality: their use in demography and acturial science
(Athens University of Economics and Business, 1998) Lytrokapi, Anna; Athens University of Economics and Business, Department of Statistics; Kostaki, Anastasia
Thesis - Athens University of Economics and Business. Postgraduate, Department of Statistics
Psychological responses of COVID 19 pandemic of students in Greek universities
(2021) Glotsou, Anna Maria; Γκλώτσου, Άννα-Μαρία; Athens University of Economics and Business, Department of Statistics; Psarakis, Stelios; Kyriakidis, Epaminondas; Kostaki, Anastasia
February 26thin 2020,was the day when the virus that had infected almost the whole world, now known as COVID-19(Corona Virus), began to spread in Greece. At the beginning of March, the first strict measures were taken, which led the Greek people to a state of confinement that lasted for almost two months. As a result, many students were trapped in the city where they were staying and due to the measures that were taken for the protection of the public health, they were for-bitten to go out of their residence unless it was necessary and were isolated from friends and relatives. Some of them experienced the confinement all alone.However, it was expected that this condition would have a direct impact on their psychology,and this could potentially lead to depression. Between the 13thof May and the 12thof June 2020, questionnaires were dispensed to the students of Greece, which were related to general information about their personal life, the conditions that prevailed during the confinement, the various ways of entertainment, as well as questions related to their emotions and mental state. The target of this paper is to obtain an overview of the situation experienced by the students in order to examine whether the prevailing conditions contributed to the formation oft heir psychology, mental health and eventually to investigate in which way these all circumstances could contribute to the provocation of depression. In order to draw conclusions, various descriptive measures (position measures, diagrams), tests for the relationship of variables (χ2test, t-test, ANOVA, Kruskal Wallis) will be estimated, as well as models will be created so to find out which factors possibly are contributing to the provocation of depression among the students. More specifically, depression will be considered as self-reported depression, which will be a quantitative variable based on eight item scaled variables, according to the Center of Epidemiological Studies-Depression (CES -D). However, it is not a clinical measure as it does not include all the variables that contribute to depression.
Statistical methodology for estimating patterns of demographic phenomena
Peristera, Paraskevi M.; Περιστερά, Παρασκευή Μ.; Athens University of Economics and Business, Department of Statistics; Kostaki, Anastasia
The main objective of this work is to develop statistical methodology for estimating patterns of demographic phenomena. The main topics are briefly described below. Several techniques have been proposed in the literature for graduating mortality data. An innovative topic for graduating mortality data is the use of kernel regression estimators. In this work we provide an analytical evaluation of the alternative kernel estimators as tools for graduating mortality data. In order to evaluate the efficiency and accuracy of kernel techniques for graduation purposes, we apply these to several empirical data sets and compare the results with those of classical graduation techniques. The interesting finding is that the Gasser-Muller estimator proves more adequate compared to other kernel-type regression estimators. The study of sex mortality differences is a topic that has been widely studied. Particular emphasis is given to the causes that provoke these differences as well as the evolution of the sex mortality gap through time. The gap between the mortality patterns of the two sexes significantly varies through ages exhibiting a pattern that became typical since the middle of the 20th century. This pattern consists of two humps, the first around age 20 and the second around age 60. However the literature lacks a model that captures this pattern. Therefore in this work a parametric model is developed for estimating the age-specific pattern of sex mortality differences over the whole age range. This model is useful for many purposes. In fact, such a model, eliminating the random variations of empirical data can serve for simplifying comparisons of the mortality patterns between sexes. Moreover, capturing the pattern into a limited set of parameter values, it can simplify comparisons of the shape and the severity of the mortality deviations of the two sexes, through space and time.In order to evaluate the performance of such a parametric representation, we fit the model to a variety of empirical data sets. The results suggest that a widening of the sex mortality differential is reflected through time. In addition in recent years, the modes of the two humps tend to move to higher ages, except of some Baltic and East-European countries. It is of course interesting to further investigate the reasons for which this sex mortality widening occurs in recent years.Furthermore, a parametric model is developed in order to graduate the age-specific patterns of fertility and nuptiality. Considering fertility, it is known that the age-specific fertility pattern has a typical shape common in all human populations through years. In recent years, a distorted fertility pattern is observed for the UK, the USA and Ireland characterised by a second hump at earlier ages. As expected, the existing models are unable to estimate the new shape of the fertility pattern and therefore the use of more appropriate representations is required. The model proposed here is able to capture both the old and the new distorted fertility pattern. In this work, this model is utilised for estimating the age-specific fertility pattern the age-specific parity rates of several European populations. The results reveal that the patterns of early-age fertility, previously confined to a few mostly English-speaking countries, are now more widely distributed in Europe. Another finding is that for countries with enhanced early-age fertility the pattern of first births also exhibits a strongly intense hump in younger ages and even stronger than the pattern of total fertility. Furthermore the fertility pattern of the USA when differentiated by the ethnicity of the mother is quite heterogeneous. These facts provide a strong evidence of heterogeneity in the female populations associated not only to the marital status, race and birth order but also to the educational level, social and economic status as well as the religiosity of the mothers. The model proposed is also adequate for estimating the age-specific pattern of nuptiality. To evaluate its adequacy, comparisons with existing models are provided, for several populations through time. The results indicate that these models generally show deviations between the empirical and estimated rates at the tails of the first-marriage distribution for the majority of populations. This might be an indication of heterogeneity in these populations and therefore, as shown here, mixture modes are more adequate for fitting the nuptiality pattern of modern populations.
Statistical methods for forecasting mortality
(Athens University of Economics and Business, 06-2012) Korakaki, Kyriaki; Kostaki, Anastasia
Thesis - Athens University of Economics and Business. Postgraduate, Department of Statistics