A partial dependence factorial analysis to deal with selection bias in observational studis

This thesis presents a creative and practical approach to dealing with the problem of selection bias. Selection bias may be the most important vexing problem in program evaluation or in any line of research that attempts to assert causality. Some of the greatest minds in economics and statistics have scrutinized the problem of selection bias, with the resulting approaches – Rubin’s Potential Outcome Approach(Rosenbaum and Rubin,1983; Rubin, 1991,2001,2004) or Heckman’s Selection model (Heckman, 1979) – being widely accepted and used as the best fixes. These solutions to the bias that arises in particular from self selection are imperfect, and many researchers, when feasible, reserve their strongest causal inference for data from experimental rather than observational studies. The innovative aspect of this thesis is to propose a data transformation that allows measuring and testing in an automatic and multivariate way the presence of selection bias. The approach involves the construction of a multi-dimensional conditional space of the X matrix in which the bias associated with the treatment assignment has been eliminated. Specifically, we propose the use of a partial dependence analysis of the X-space as a tool for investigating the dependence relationship between a set of observable pre-treatment categorical covariates X and a treatment indicator variable T, in order to obtain a measure of bias according to their dependence structure. The measure of selection bias is then expressed in terms of inertia due to the dependence between X and T that has been eliminated. Given the measure of selection bias, we propose a multivariate test of imbalance in order to check if the detected bias is significant, by using the asymptotical distribution of inertia due to T (Estadella et al. 2005) , and by preserving the multivariate nature of data. Further, we propose the use of a clustering procedure as a tool to find groups of comparable units on which estimate local causal effects, and the use of the multivariate test of imbalance as a stopping rule in choosing the best cluster solution set. The method is non parametric, it does not call for modeling the data, based on some underlying theory or assumption about the selection process, but instead it calls for using the existing variability within the data and letting the data to speak. The idea of proposing this multivariate approach to measure selection bias and test balance comes from the consideration that in applied research all aspects of multivariate balance, not represented in the univariate variable- by-variable summaries, are ignored. The first part contains an introduction to evaluation methods as part of public and private decision process and a review of the literature of evaluation methods. The attention is focused on Rubin Potential Outcome Approach, matching methods, and briefly on Heckman’s Selection Model. The second part focuses on some resulting limitations of conventional methods, with particular attention to the problem of how testing in the correct way balancing. The third part contains the original contribution proposed , a simulation study that allows to check the performance of the method for a given dependence setting and an application to a real data set. Finally, we discuss, conclude and explain our future perspectives.

Essays on the Empirical Analysis of Economic and Political Development in Sub-Saharan Africa

In Sub-Saharan Africa, non-democratic events, like civil wars and coup d'etat, destroy economic development. This study investigates both domestic and spatial effects on the likelihood of civil wars and coup d'etat. To civil wars, an increase of income growth is one of common research conclusions to stop wars. This study adds a concern on ethnic fractionalization. IV-2SLS is applied to overcome causality problem. The findings document that income growth is significant to reduce number and degree of violence in high ethnic fractionalized countries, otherwise they are trade-off. Income growth reduces amount of wars, but increases its violent level, in the countries with few large ethnic groups. Promoting growth should consider ethnic composition. This study also investigates the clustering and contagion of civil wars using spatial panel data models. Onset, incidence and end of civil conflicts spread across the network of neighboring countries while peace, the end of conflicts, diffuse only with the nearest neighbor. There is an evidence of indirect links from neighboring income growth, without too much inequality, to reduce the likelihood of civil wars. To coup d'etat, this study revisits its diffusion for both all types of coups and only successful ones. The results find an existence of both domestic and spatial determinants in different periods. Domestic income growth plays major role to reduce the likelihood of coup before cold war ends, while spatial effects do negative afterward. Results on probability to succeed coup are similar. After cold war ends, international organisations seriously promote democracy with pressure against coup d'etat, and it seems to be effective. In sum, this study indicates the role of domestic ethnic fractionalization and the spread of neighboring effects to the likelihood of non-democratic events in a country. Policy implementation should concern these factors.

Financial Development and Economic Growth: Time Series Evidence from Albania

The main objective of this thesis is to explore the short and long run causality patterns in the finance – growth nexus and finance-growth-trade nexus before and after the global financial crisis, in the case of Albania. To this end we use quarterly data on real GDP, 13 proxy measures for financial development and the trade openness indicator for the period 1998Q1 – 2013Q2 and 1998Q1-2008Q3. Causality patterns will be explored in a VAR-VECM framework. For this purpose we will proceed as follows: (i) testing for the integration order of the variables; (ii) cointegration analysis and (iii) performing Granger causality tests in a VAR-VECM framework. In the finance-growth nexus, empirical evidence suggests for a positive long run relationship between finance and economic growth, with causality running from financial development to economic growth. The global financial crisis seems to have not affected the causality direction in the finance and growth nexus, thus supporting the finance led growth hypothesis in the long run in the case of Albania. In the finance-growth-trade openness nexus, we found evidence for a positive long run relationship the variables, with causality direction depending on the proxy used for financial development. When the pre-crisis sample is considered, we find evidence for causality running from financial development and trade openness to economic growth. The global financial crisis seems to have affected somewhat the causality direction in the finance-growth-trade nexus, which has become sensible to the proxy used for financial development. On the short run, empirical evidence suggests for a clear unidirectional relationship between finance and growth, with causality mostly running from economic growth to financial development. When we consider the per-crisis sub sample results are mixed, depending on the proxy used for financial development. The same results are confirmed when trade openness is taken into account.

Timeliness, efficiency and public procurement in hip surgery in Italy

Considering different perspectives, the scope of this thesis is to investigate how to improve healthcare resources allocation and the provision efficiency for hip surgeries, a resource-intensive operation, among the most frequently performed on the elderly, with a trend in volume that is increasing in years due to population aging. Firstly, the effect of Time-To-Surgery (TTS) on mortality for hip fracture patients is investigated. The analysis attempts to account for TTS endogeneity due to the inability to fully control for variables affecting patient delay – e.g. patient severity. Exploiting an instrumental variable model, where being admitted on Friday or Saturday predicts longer TTS, findings show exogenous TTS does not have a significant effect on mortality. Thus suggesting surgeons prioritize patients effectively, neutralizing the adverse impact of longer TTS. Then, the volume-outcome relation for total hip replacement surgery is analyzed, seeking to account for selective referral, which may be present in elective surgery context, and induce reverse causality issue in the volume-outcome relation. The analysis employs a conditional choice model where patient travel distance from all regions' hospitals is used as a hospital choice predictor. Findings show the exogenous hospital volume significantly decreases adverse outcomes probability, especially in the short run. Finally, the change in public procurement design enforced in the Romagna LHA (Italy) is exploited to assess its impact on hip prostheses cost, surgeons' implant choice, and patient health outcomes. Hip prostheses are the major cost-driver of hip replacement surgeries, hence it is crucial to design the public tender such that implant prices are minimized, but cost-containment policies have to be weighted with patient well-being. Evidence shows that a cost reduction occurred without a significant surgeons’ choices impact. Positive or no effect of surgeons specialization is found on patients outcomes after the new procurement introduction.

Diagnostica anatomopatologica nelle malattie infettive del suino: co-localizzazione agenti patogeni e lesioni, limiti ed utilità

Il processo di intensificazione dell’allevamento suinicolo ha comportato conseguenze importanti dal punto di vista sanitario. Nei moderni sistemi di produzione la salute è compromessa a causa di malattie multifattoriali e l'identificazione degli agenti casuali di malattia non può basarsi esclusivamente sul risultato di un unico test diagnostico. La maggior parte dei patogeni è ubiquitaria, quindi l'identificazione di un agente infettivo suggerisce solo una possibile diagnosi. La patologia gioca un ruolo chiave nel determinare la causalità nella diagnosi delle malattie infettive. L'istopatologia consente di correlare la presenza dell'agente infettivo con le lesioni microscopiche caratteristiche: in base alla disponibilità di anticorpi o sonde per rilevare rispettivamente la presenza di antigeni o del genoma di virus o batteri nelle sezioni istologiche, l'istopatologia è utile per determinare il nesso causale con la malattia. L'introduzione di questa tesi è incentrata sulla discussione del ruolo della patologia nelle malattie infettive del suino, seguita dalla descrizione del corretto campionamento per l'esame istopatologico nella diagnosi delle malattie respiratorie, enteriche e riproduttive. La sezione sperimentale riporta i risultati delle indagini diagnostiche in patologia suina condotte per definire il ruolo di PCV2 e PRRSV nel complesso delle malattie respiratorie del suino (PRDC)(capitolo 2) e per definire un percorso diagnostico per l'enteropatia proliferativa suina causata da Lawsonia intracellularis ai fini dell'implementazione dell'accuratezza diagnostica (capitolo 3). Il capitolo 4 descrive l'uso dell'istopatologia come componente metodologica essenziale per valutare l'efficacia protettiva dei vaccini negli studi di challenge vaccinazione-infezione. Il capitolo descrive i risultati ottenuti in due successive sperimentazioni per valutare l'efficacia di due vaccini sperimentali per PCV2, il primo formulato con ceppo PCV2b inattivato somministrato a differenti dosi di antigene; il secondo formulato con antigene "Cap" ricombinante non assemblato in particelle simil-virali (VLP).

Assessing brain connectivity through electroencephalographic signal processing and modeling analysis

Brain functioning relies on the interaction of several neural populations connected through complex connectivity networks, enabling the transmission and integration of information. Recent advances in neuroimaging techniques, such as electroencephalography (EEG), have deepened our understanding of the reciprocal roles played by brain regions during cognitive processes. The underlying idea of this PhD research is that EEG-related functional connectivity (FC) changes in the brain may incorporate important neuromarkers of behavior and cognition, as well as brain disorders, even at subclinical levels. However, a complete understanding of the reliability of the wide range of existing connectivity estimation techniques is still lacking. The first part of this work addresses this limitation by employing Neural Mass Models (NMMs), which simulate EEG activity and offer a unique tool to study interconnected networks of brain regions in controlled conditions. NMMs were employed to test FC estimators like Transfer Entropy and Granger Causality in linear and nonlinear conditions. Results revealed that connectivity estimates reflect information transmission between brain regions, a quantity that can be significantly different from the connectivity strength, and that Granger causality outperforms the other estimators. A second objective of this thesis was to assess brain connectivity and network changes on EEG data reconstructed at the cortical level. Functional brain connectivity has been estimated through Granger Causality, in both temporal and spectral domains, with the following goals: a) detect task-dependent functional connectivity network changes, focusing on internal-external attention competition and fear conditioning and reversal; b) identify resting-state network alterations in a subclinical population with high autistic traits. Connectivity-based neuromarkers, compared to the canonical EEG analysis, can provide deeper insights into brain mechanisms and may drive future diagnostic methods and therapeutic interventions. However, further methodological studies are required to fully understand the accuracy and information captured by FC estimates, especially concerning nonlinear phenomena.

Comparative constitutional design for divided societies: a model to explain vonstitutional asymmetries

Comparative studies on constitutional design for divided societies indicate that there is no magic formula to the challenges that these societies pose, as lots of factors influence constitutional design. In the literature on asymmetric federalism, the introduction of constitutional asymmetries is considered a flexible instrument of ethnic conflict resolution, as it provides a mixture of the two main theoretical approaches to constitutional design for divided societies (i.e., integration and accommodation). Indeed, constitutional asymmetries are a complex and multifaceted phenomenon, as their degree of intensity can vary across constitutional systems, and there are both legal and extra-legal factors that may explain such differences. This thesis argues that constitutional asymmetries provide a flexible model of constitutional design and aims to explore the legal factors that are most likely to explain the different degrees of constitutional asymmetry in divided multi-tiered systems. To this end, the research adopts a qualitative methodology, i.e., Qualitative Comparative Analysis (QCA), which allows an understanding of whether a condition or combination of conditions (i.e., the legal factors) determine the outcome (i.e., high, medium, low degree of constitutional asymmetry, or constitutional symmetry). The QCA is conducted on 16 divided multi-tiered systems, and for each case, the degree of constitutional asymmetry was analyzed by employing standardized indexes on subnational autonomy, allowing for a more precise measure of constitutional asymmetry than has previously been provided in the literature. Overall, the research confirms the complex nature of constitutional asymmetries, as the degrees of asymmetries vary substantially not only across systems but also within cases among the dimensions of subnational autonomy. The outcome of the Qualitative Comparative Analysis also confirms a path of complex causality since the different degrees of constitutional asymmetry always depend on several legal factors, that combined produce a low, medium, or high degree of constitutional asymmetry or, conversely, constitutional symmetry.

Idee, Natur, Geist. Schellings allgemeine Ontologie des Wirklichen nach der "Darstellung des Naturprocesses" von 1843/44

Ziel dieser Dissertation ist es, die grundlegenden philosophisch-theoretischen Implikationen von Schellings letzter systematischer Darlegung seiner Naturphilosophie nach dem Berliner Textfragment von 1843/44, der “Darstellung des Naturprocesses”, zu untersuchen. Angesichts der sich zwischen den 1830 und den 1860 Jahren in Berlin abzeichnenden neuen intellektuellen Tendenzen und der Entwicklungen in den Naturwissenschaften legt Schelling hier die Grundlagen für eine allgemeine Ontologie des Wirklichen in kritischer Auseinandersetzung mit Kants transzendentalem Idealismus. Innerhalb des systematischen Horizonts der "apriorischen Vernunftwissenschaft" oder "negativen Philosophie" stellt er im ersten Teil seines Werkes die Prinzipien fest, die die „Idee des Existierenden“ ausmachen, und beschreibt die rationale Operation, die durchgeführt werden muss, um zum Gedanken einer „Welt außer der Idee“ zu gelangen. Die philosophisch-systematischen Annahmen, die mit dem Übergang von der bloßen Idee des Existierenden zum Gedanken der außeridealen Welt verbunden sind, werden im ersten Kapitel dieser Dissertation untersucht. Im zweiten Teil seines Werkes definiert Schelling durch eine detaillierte Analyse von Kants Transzendentalen Ästhetik den Raum als diejenige Form, in der uns die Existenzen als voneinander getrennt vorstellen lassen. Obwohl der Zeitbegriff von Schelling nur am Rande behandelt wird, trägt er zusammen mit dem Raum dazu bei, die erste ontologische Grundstruktur der außeridealen Welt zu definieren. Die Analyse von Schellings Konzeption der raumzeitlichen Grundstruktur der außeridealen Welt stellt das Thema des zweiten Kapitels dieser Dissertation dar. Schließlich bestimmt Schelling im dritten Teil seines Werkes die Finalität als diejenige Kausalitätsform, die es ermöglicht, die außerideale Welt als einen werdenden Kontext zu verstehen, dessen Entwicklungsstufen die siderische Welt, die unorganische Welt und die organische Welt sind. Die Schelling‘sche Definition der Teleologie der Natur als zweite ontologische Grundstruktur der außeridealen Welt ist das Thema des dritten Kapitels dieser Dissertation.