We investigate stability for certain geometric and functional inequalities and address the regularity of the free boundary for a problem arising in optimal transport theory. More specifically, stability estimates are obtained for the relative isoperimetric inequality inside convex cones and the Gaussian log-Sobolev inequality for a two parameter family of functions. Thereafter, away from a ``small" singular set, local C^{1,\alpha} regularity of the free boundary is achieved in the optimal partial transport problem. Furthermore, a technique is developed and implemented for estimating the Hausdorff dimension of the singular set. We conclude with a corresponding regularity theory on Riemannian manifolds. / text

The aim of the dissertation research is to identify the features of Lithuanian educational system affecting social inequality in the direction to stimulation or reduction. The role of education regarding social inequality is one of the most controversial objects of scientific discussions. Is education an effective foundation of society emancipation, equal wrights and social mobility, or on the contrary, it enhances the current structure of the society? It is an open question for scientific discussions. Researches show that education connect in itself two contradictional elements: on the one hand, it is one of the most widely acknowledged social “elevators” helping to rise by the steps of social status, on the other hand, education also has the function of mobility control. Many researches revealed that correlations between education and social status of a man in different countries are different. Thus the role of education with respect to social inequality is directly connected with social, economical and cultural specificity of the country. The dissertation research employs tools of qualitative research, and they allow describe the relationship between education and social inequality as well as identify concrete features of education system that diminish or/and increase social inequality in specific socio-cultural and economic conditions in Lithuania. The composition of the research sample includes the levels of national politics and self-government, institution (education... [to full text] / Disertacijos tikslas - identifikuoti Lietuvos švietimo sistemos bruožus, paveikiančius socialinę nelygybę jos skatinimo arba redukavimo kryptimi. Švietimo vaidmuo socialinės nelygybės atžvilgiu yra vienas prieštaringiausių mokslinių diskusijų objektas. Ar švietimas iš tiesų yra efektyvus visuomenės emancipacijos, lygių galimybių bei socialinio mobilumo pamatas, ar atvirkščiai – linkęs išlaikyti esamą socialinę visuomenės struktūrą dar šiandien yra atviras mokslinėms diskusijoms klausimas. Tyrimai rodo, jog švietimas savyje apjungia du prieštaraujančius pradus – viena vertus, jis yra vienas plačiausiai pripažintų socialinių „liftų“, padedančių pakilti socialinio statuso laipteliais aukštyn kita vertus, švietimas atlieka ir mobilumo kontrolės funkciją. Dauguma tyrimų atskleidė, jog skirtingose šalyse koreliacijos tarp išsimokslinimo ir užimamo statuso yra skirtingos. Taigi, švietimo vaidmuo socialinės nelygybės atžvilgiu, tiesiogiai siejasi su šalies socialiniu, ekonominiu ir kultūriniu specifiškumu. Disertaciniame tyrime taikytos kokybinio tyrimo priemonės kaip tik ir leido aprašyti švietimo ir socialinės nelygybės sąryšius bei identifikuoti konkrečius švietimo sistemos bruožus, mažinančius ir/arba didinančius socialinę nelygybę specifinėmis Lietuvos sociokultūrinėmis ir ekonominėmis sąlygomis. Sudarant tyrimo imtį kryptingai buvo apjungtas nacionalinės politikos ir savivaldos lygmuo, institucinis (švietimos ir kitos socialinės institucijos) bei individualus lygmuo (mokytojai... [toliau žr. visą tekstą]

No description available.

Die fünf Essays dieser Dissertation behandeln Themen aus dem Bereich der Entwicklungs- und Umweltökonomie. Alle Essays analysieren wie die Produktion von CO2 Emissionen beeinflusst oder reguliert werden kann. Das Treibhausgas CO2 ist eine der größten Externalitäten der Geschichte menschlicher Entwicklung. Die einzelnen Essays zeigen wie lokale Klimaveränderungen das menschliche Wohlbefinden beeinflussen und welche monetären Kosten mit einem Anstieg der Durchschnittstemperatur in Lateinamerika verbunden sind. Außerdem betrachten die Essays Hauptdeterminanten von CO2 Emissionen auf haushalts- oder nationalem Niveau, und bestimmen den Erfolg aktueller Klimapolitik um CO2 Emissionen zu reduzieren. Das letzte Essay betrachtet die momentane und zukünftige Verteilung von CO2 Emissionen wenn verschiedene Politikszenarien realisiert werden würden. Das erste Essay befasst sich mit dem Effekt von klimatischen Veränderungen auf das Wohlfahrtsniveau in Lateinamerika. Als Wohlfahrtsmaß kommen dabei subjektive Selbstaussagen zum Einsatz. Subjektive Wohlfahrt erfasst nicht nur Veränderungen im Einkommen, sondern auch Veränderungen in anderen Lebensbereichen wie dem Zugang zu Bildung oder Gesundheitseinrichtungen. Generell kommt die Studie zu dem Schluss, dass eine Temperatur im Bereich von 20 Grad Celsius und Niederschlag bis 247mm optimal sind. Höhere monatliche Durchschnittstemperaturen oder Niederschläge sind mit Wohlfahrtsverlusten verbunden. Eine globale Erwärmung von mehr als 2 Grad Celsius wird mit Wohlfahrtsverlusten in Lateinamerika einhergehen. Das zweite Essay analysiert Haushaltsemissionen in Form des Kohlenstoff-Fußabdrucks in Indien. Dabei liegt das Augenmerk auf dem Effekt von Einkommenswachstum und sozio- ökonomischen Veränderungen innerhalb der Haushalte. Ein höheres Haushaltseinkommen führt zu einem stärkeren Konsumverhalten aber gleichzeitig auch zu weniger CO2- intensiven Konsummustern. Dennoch kann der Mehrkonsum an CO2-armen Gütern, wie zum Beispiel Bildung, den Anstieg der Haushaltsemissionen, aufgrund höheren Einkommens, nicht kompensieren. Das dritte Essay betrachtet in wie fern aktuelle internationale Klimapolitik einen Einfluss auf CO2 Emissionen genommen hat. Dabei zeigt sich, dass Länder, welche Verpflichtungen im Rahmen des Kyoto Protokolls eingegangen sind, im Durschnitt 6.5% weniger CO2 emittiert haben, als vergleichbare Länder mit ähnlichem Einkommens- und Bevölkerungswachstum aber ohne Verpflichtungen. Das vierte Essay geht auf die Hauptdeterminante des CO2 Emissionswachstums ein, nämlich Einkommen. Dabei wird aber nicht nur der Effekt von Veränderungen im Einkommen, sondern auch der Effekt von Veränderungen in der Einkommensverteilung auf CO2 Emissionen untersucht. Einkommensungleichheit wirkt sich abhängig vom gegenwertigen Ungleichheitsniveau auf CO2 Emissionen aus. Für Länder mit einer hohen Einkommensungleichheit ist der Effekt positiv, das heißt mit sinkender Einkommensungleichheit sinken CO2 Emissionen. Für Länder mit niedriger Ungleichheit ist der Effekt negativ. Ein weiterer Abbau der Einkommensungleichheit würde dort mit steigenden CO2 Emissionen einhergehen. Das fünfte Essay befasst sich mit der globalen Verteilung von pro Kopf CO2 Emissionen. Dabei geht es darum inwiefern der Energiemix und der sektorale Aufbau einzelner Volkswirtschaften zu dieser ungleichen Verteilung von pro Kopf CO2 Emissionen beigetragen haben. Der Abbau schwerer Industrie in OECD Ländern und der verstärkte Einsatz von Kohle in nicht-OECD Ländern haben dabei zu einem Rückgang der globalen Ungleichheit in CO2 Emissionen geführt. Langfristig gesehen kann es sein, dass die Emissionsungleichheit ab 2040 wieder steigen wird. Jedes Essay trägt in seinem Feld zur betreffenden Literatur bei. Die Essays analysieren wie jegliche ökonomische Aktivität (hauptsächlich Konsum) CO2 Emissionen verursachen, welche wiederum für Veränderungen im Klima verantwortlich gemacht werden. Diese Veränderungen im Klima gehen mit lokalen Wohlfahrtsverlusten einher. Nationale Politikmaßnahmen wie zum Beispiel Maßnahmen zur Einkommensumverteilung können einen ambivalenten Einfluss auf CO2 Emissionen haben. Politikmaßnahmen um das Konsumverhalten und Konsummuster zu beeinflussen könnten ein effizientes Mittel zur Regulierung von CO2 Emissionen in reichen Ländern darstellen. Generell könnten internationale Klimapolitikmaßnahmen nationale Politikmaßnahmen katalysieren.

The thesis considers two pertinent questions about poverty in Fiji. One is about the accuracy of the poverty measures calculated by the concerned organisations and this relates to the use of equivalence scales and the general style of analysis. The other more intricate question is the disregard for poverty due to intra-family distribution asymmetries. Such miscalculations of poverty arise due use of average household per capita expenditure to represent consumption. This research attempts to answer the question of whether the tendency to underestimate the incidence of poverty by disregarding intra-family inequality is significant. Furthermore, it attempts to determine the causes of these inequalities. The issue is whether the classical method of data analysis (using the family as a unit) is the ideal way of analysing poverty and distribution in societies where large family structures exist and government relief remains minimal. To determine the household inequalities, household expenditures have been disaggregated into individualised expenditures. The individualised consumption expenditure is analysed and compared with the outcomes of aggregate household expenditure data. The analysis provides overwhelming evidence for underestimation of poverty when household consumption expenditures are used.

The thesis considers two pertinent questions about poverty in Fiji. One is about the accuracy of the poverty measures calculated by the concerned organisations and this relates to the use of equivalence scales and the general style of analysis. The other more intricate question is the disregard for poverty due to intra-family distribution asymmetries. Such miscalculations of poverty arise due use of average household per capita expenditure to represent consumption. This research attempts to answer the question of whether the tendency to underestimate the incidence of poverty by disregarding intra-family inequality is significant. Furthermore, it attempts to determine the causes of these inequalities. The issue is whether the classical method of data analysis (using the family as a unit) is the ideal way of analysing poverty and distribution in societies where large family structures exist and government relief remains minimal. To determine the household inequalities, household expenditures have been disaggregated into individualised expenditures. The individualised consumption expenditure is analysed and compared with the outcomes of aggregate household expenditure data. The analysis provides overwhelming evidence for underestimation of poverty when household consumption expenditures are used.

This study documents how organizational strategies underlying college counseling departments modify counselors’ ability to perform their academic and college advising duties. To examine this, fifteen semi-structured, in-depth interviews with public high school counselors in Southern California were conducted. A district’s commitment to college access and opportunity, as well as parents’ expectations for maintaining a college-going culture, shaped the nature of college counseling and organizational habitus in a school. Counselors reported that access to different forms of institutional support and resources diminish or exacerbate the structural constraints known to surface in public schools. This influenced when and how counselors advised students, which offered different views of the educational opportunity structure. Additionally, findings indicate that both college preparation programs and culturally sensitive counselors work to remedy educational disparities and increase the college enrollment rates of working-class minority students.

Submitted by Guido Couto Penido Guimarães (guido.guimaraes@fgvmail.br) on 2015-05-15T19:21:28Z No. of bitstreams: 1 20150516035825408.pdf: 2427180 bytes, checksum: 7d04b8fe0a5b09cdefb5e1395b3e7dbd (MD5) / Approved for entry into archive by BRUNA BARROS (bruna.barros@fgv.br) on 2015-05-20T14:02:22Z (GMT) No. of bitstreams: 1 20150516035825408.pdf: 2427180 bytes, checksum: 7d04b8fe0a5b09cdefb5e1395b3e7dbd (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2015-06-12T20:07:15Z (GMT) No. of bitstreams: 1 20150516035825408.pdf: 2427180 bytes, checksum: 7d04b8fe0a5b09cdefb5e1395b3e7dbd (MD5) / Made available in DSpace on 2015-06-12T20:07:40Z (GMT). No. of bitstreams: 1 20150516035825408.pdf: 2427180 bytes, checksum: 7d04b8fe0a5b09cdefb5e1395b3e7dbd (MD5) Previous issue date: 2015-04-13 / In this paper we investiga te the impact of initial wealth anel impatience heterogeneities, as wcll as differential access to financia! markets on povcrty anel inequality, anel cvaluate some mechanisms that could be used to alleviate situations in which these two issues are alarming. To address our qucstion we develop a dynamic stochastic general cquilibrium modo! of educational anel savings choicc with heterogeneous agents, where individuais differ in their initial wealth anel in their discount factor. We find that, in the long run, more patient households tend to be wealthier anel more educated. However, our baseline model is not able to give as much skewness to our income distribution as it is rcquircd. We then propose a novel returns structure based on empírica! observation of heterogeneous returns to different portfolios. This modification solves our previous problem, evidencing the importance of the changes made in explaining the existing levels of inequality. Finally, we introducc two kinds of cash transfers programs- one in which receiving thc benefit is conditional on educating the household's youngster (CCTS) anel one frec of conditionalities (CTS) - in order to evaluate the impact of these programs on the variables of concern1 Wc fine! that both policies have similar qualitativo rcsults. Quantitatively, howcvcr, the CCTS outperforms its unconclitional version in all fielcls analyzecl, revealing itself to be a preferable policy.

Continuous change and development are essential aspects of evolving environments and applications, including, but not limited to, smart cities, military, medicine, nuclear reactors, self-driving cars, aviation, and aerospace. That is, the fundamental characteristics of such environments may evolve, and so cause dangerous consequences, e.g., putting people lives at stake, if no reaction is adopted. Therefore, learning systems need to apply intelligent algorithms to monitor evolvement in their environments and update themselves effectively. Further, we may experience fluctuations regarding the performance of learning algorithms due to the nature of incoming data as it continuously evolves. That is, the current efficient learning approach may become deprecated after a change in data or environment. Hence, the question 'how to have an efficient learning algorithm over time against evolving data?' has to be addressed. In this thesis, we have made two contributions to settle the challenges described above. In the machine learning literature, the phenomenon of (distributional) change in data is known as concept drift. Concept drift may shift decision boundaries, and cause a decline in accuracy. Learning algorithms, indeed, have to detect concept drift in evolving data streams and replace their predictive models accordingly. To address this challenge, adaptive learners have been devised which may utilize drift detection methods to locate the drift points in dynamic and changing data streams. A drift detection method able to discover the drift points quickly, with the lowest false positive and false negative rates, is preferred. False positive refers to incorrectly alarming for concept drift, and false negative refers to not alarming for concept drift. In this thesis, we introduce three algorithms, called as the Fast Hoeffding Drift Detection Method (FHDDM), the Stacking Fast Hoeffding Drift Detection Method (FHDDMS), and the McDiarmid Drift Detection Methods (MDDMs), for detecting drift points with the minimum delay, false positive, and false negative rates. FHDDM is a sliding window-based algorithm and applies Hoeffding’s inequality (Hoeffding, 1963) to detect concept drift. FHDDM slides its window over the prediction results, which are either 1 (for a correct prediction) or 0 (for a wrong prediction). Meanwhile, it compares the mean of elements inside the window with the maximum mean observed so far; subsequently, a significant difference between the two means, upper-bounded by the Hoeffding inequality, indicates the occurrence of concept drift. The FHDDMS extends the FHDDM algorithm by sliding multiple windows over its entries for a better drift detection regarding the detection delay and false negative rate. In contrast to FHDDM/S, the MDDM variants assign weights to their entries, i.e., higher weights are associated with the most recent entries in the sliding window, for faster detection of concept drift. The rationale is that recent examples reflect the ongoing situation adequately. Then, by putting higher weights on the latest entries, we may detect concept drift quickly. An MDDM algorithm bounds the difference between the weighted mean of elements in the sliding window and the maximum weighted mean seen so far, using McDiarmid’s inequality (McDiarmid, 1989). Eventually, it alarms for concept drift once a significant difference is experienced. We experimentally show that FHDDM/S and MDDMs outperform the state-of-the-art by representing promising results in terms of the adaptation and classification measures. Due to the evolving nature of data streams, the performance of an adaptive learner, which is defined by the classification, adaptation, and resource consumption measures, may fluctuate over time. In fact, a learning algorithm, in the form of a (classifier, detector) pair, may present a significant performance before a concept drift point, but not after. We define this problem by the question 'how can we ensure that an efficient classifier-detector pair is present at any time in an evolving environment?' To answer this, we have developed the Tornado framework which runs various kinds of learning algorithms simultaneously against evolving data streams. Each algorithm incrementally and independently trains a predictive model and updates the statistics of its drift detector. Meanwhile, our framework monitors the (classifier, detector) pairs, and recommends the efficient one, concerning the classification, adaptation, and resource consumption performance, to the user. We further define the holistic CAR measure that integrates the classification, adaptation, and resource consumption measures for evaluating the performance of adaptive learning algorithms. Our experiments confirm that the most efficient algorithm may differ over time because of the developing and evolving nature of data streams.

Submitted by ANA KARLA PEREIRA RODRIGUES (anakarla_@hotmail.com) on 2017-08-17T16:04:13Z No. of bitstreams: 1 arquivototal.pdf: 699730 bytes, checksum: d48ddf5357572db4dac922761f91c532 (MD5) / Made available in DSpace on 2017-08-17T16:04:13Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 699730 bytes, checksum: d48ddf5357572db4dac922761f91c532 (MD5) Previous issue date: 2016-04-04 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / In this study we show two classical inequalities, namely Bohnenblust-Hille inequality and Hardy-Littlewood inequality. The rst one, conceived as a tool for the study of problems related to Dirichlet series, is a generalization of Littlewood`s 4/3 inequality to multilinear forms. The second is a generalization of Bohnenblust-Hille inequality, produced by the replacement of c0 with lp. / No presente trabalho abordaremos duas desigualdades cl assicas, a saber, a Desigualdade de Bohnenblust-Hille e a Desigualdade de Hardy- Littlewood. A primeira, surgiu como ferramenta para o estudo de problemas relacionados a s eries de Dirichlet e e uma generaliza c~ao para formas multilineares da Desigualdade 4/3 de Littlewood. A segunda consiste de uma generaliza c~ao da Desigualdade de Bohnenblust-Hille, produzida pela substituição de c0 por lp.