Survival analysis is an essential statistics and machine learning field in various critical applications like medical research and predictive maintenance. In these domains understanding models' predictions is paramount. While machine learning techniques are increasingly applied to enhance the predictive performance of survival models, they simultaneously sacrifice transparency and explainability.  Survival models, in contrast to regular machine learning models, predict functions rather than point estimates like regression and classification models. This creates a challenge regarding explaining such models using the known off-the-shelf machine learning explanation techniques, like Shapley Values, Counterfactual examples, and others.    Censoring is also a major issue in survival analysis where the target time variable is not fully observed for all subjects. Moreover, in predictive maintenance settings, recorded events do not always map to actual failures, where some components could be replaced because it is considered faulty or about to fail in the future based on an expert's opinion. Censoring and noisy labels create problems in terms of modeling and evaluation that require to be addressed during the development and evaluation of the survival models. Considering the challenges in survival modeling and the differences from regular machine learning models, this thesis aims to bridge this gap by facilitating the use of machine learning explanation methods to produce plausible and actionable explanations for survival models. It also aims to enhance survival modeling and evaluation revealing a better insight into the differences among the compared survival models. In this thesis, we propose two methods for explaining survival models which rely on discovering survival patterns in the model's predictions that group the studied subjects into significantly different survival groups. Each pattern reflects a specific survival behavior common to all the subjects in their respective group. We utilize these patterns to explain the predictions of the studied model in two ways. In the first, we employ a classification proxy model that can capture the relationship between the descriptive features of subjects and the learned survival patterns. Explaining such a proxy model using Shapley Values provides insights into the feature attribution of belonging to a specific survival pattern. In the second method, we addressed the "what if?" question by generating plausible and actionable counterfactual examples that would change the predicted pattern of the studied subject. Such counterfactual examples provide insights into actionable changes required to enhance the survivability of subjects. We also propose a variational-inference-based generative model for estimating the time-to-event distribution. The model relies on a regression-based loss function with the ability to handle censored cases. It also relies on sampling for estimating the conditional probability of event times. Moreover, we propose a decomposition of the C-index into a weighted harmonic average of two quantities, the concordance among the observed events and the concordance between observed and censored cases. These two quantities, weighted by a factor representing the balance between the two, can reveal differences between survival models previously unseen using only the total Concordance index. This can give insight into the performances of different models and their relation to the characteristics of the studied data. Finally, as part of enhancing survival modeling, we propose an algorithm that can correct erroneous event labels in predictive maintenance time-to-event data. we adopt an expectation-maximization-like approach utilizing a genetic algorithm to find better labels that would maximize the survival model's performance. Over iteration, the algorithm builds confidence about events' assignments which improves the search in the following iterations until convergence. We performed experiments on real and synthetic data showing that our proposed methods enhance the performance in survival modeling and can reveal the underlying factors contributing to the explainability of survival models' behavior and performance.

Functional Theory, the prevailing perspective on the function of counterfactual thinking, posits that the primary purpose of this form of mental simulation is to prepare individuals for the future. However, recent findings have presented challenges to this dominant view. The debate on this topic has recently centered around the possibility that these contradictory results may have arisen from the use of tasks that are inadequate to observe the preparatory function of counterfactuals. Moreover, it has been stressed the importance of considering also more spontaneous (or, at least, more intrinsically motivated) instances of counterfactual thinking when reflecting over its function. In this thesis, Experiments 1 to 3 investigated questions related to the content of counterfactual modifications and their beneficial effects on future performance, utilizing a novel task designed specifically to address previous limitations advanced by the proponents of the Functional Theory. Nonetheless, most of our results did not align with what would have been expected if counterfactual thoughts were produced mainly in a preparatory fashion. Experiments 4 to 6, instead, explored individuals’ inclination to look for non-instrumental counterfactual information, which pertains to information about the outcome of forgone options that does not contribute to improving future outcomes. This type of information seeking, involving the comparison between an actual and a forgone outcome, can represent a proxy of an underlying, genuine process of counterfactual analysis, and thus provides valuable insights for the debate on the function of counterfactual thinking. Indeed, results indicated that individuals readily look for counterfactual information even when it cannot serve any preparatory goal, challenging the notion that the consideration of alternatives to past events is strictly tied to the presence of such goals. These findings, along with prior research, raise questions about the extent to which counterfactual thoughts are produced to prepare for the future, prompting a reevaluation of its underlying functions.

No description available.

No description available.

Hur har vi kunskap om vad som är möjligt? Enligt vad som kan betraktas som det traditionella svaret till den frågan, har vi kunskap om modalitet via föreställningsbarhet. Vi föreställer oss ting och tar sedan detta som bevis för möjlighet. Denna uppsats kommer att undersöka tre invändningar till detta svar angående hur vi har kunskap om möjlighet. Vi kommer sedan att överväga Williamsons förmodan: att vår kognitiva kapacitet för att hantera kontrafaktiska konditionaler bär med sig den kognitiva kapaciteten för oss att även hantera metafysisk modalitet (2007, 136), och undersöka om denna förmodan undviker dessa invändningar. Det kommer här att argumenteras att Williamson’s förmodan undviker två av invändningarna och att den inte tycks kunna svara på den sista invändningen. Det kommer även att argumenteras att en invändning mot Williamson’s förmodan ser ut att vara särskilt problematisk, och att det inte är klart att Williamson’s förmodan är i någon bättre position än den negativa föreställningsbarhets vyn. / How do we have knowledge of what is possible? On what could be considered as the traditional response to this question, we have knowledge of modality by conceivability. We conceive of things and on the basis take this as evidence for possibility. This thesis will consider three objections to this response of how we have knowledge of possibility. We will then consider Williamson’s conjecture: that our cognitive capacity to handle counterfactual conditionals carries the cognitive capacity for us to also handle metaphysical modality (2007, 136), and see if this conjecture avoids these objections. It will be argued that Williamson’s conjecture avoids two of the objections and that it does not seem to have a response to the last objection. It will also be argued that one objection to Williamson’s conjecture seems particularly problematic, and that it is not so clear that Williamson’s conjecture is any better off than the negative conceivability view.

[pt] Usando uma recém publicada base de dados pelo Banco Central do Brazil (BCB), estimamos os efeitos de intervenções cambiais realizadas entre 1999 e 2023. Em primeiro lugar, utilizamos um VAR estrutural em frequência diárias, identificado por meio de um instrumento baseado nos horários de anúncio das interveções. Estima-se que as intervenções são capazes de afetar o nível do Real por um período de 20 dias úteis, em 0.24 p.p. a cada bilhão de dólares empregados. / [en] Benefiting from a novel dataset published by the Central Bank of Brazil (BCB) we estimate the effects of FX Interventions from 1999 to 2023. We first employ a structural VAR with daily frequency identified with an instrument based on the timing of BCB announcements. Interventions are found to be effective in changing the USDBRL level over a period of 20 working days by 0.24 p.p. for each 1USD billion employed. We then implement an Artificial Counterfactual (ArCo) approach to each intervention episode separating them by side and instrument. Compared to SVAR interventions are found to be more effective although with smaller statistical significance. Spot Interventions are more effective than Swaps. We find no effects of interventions over the shortterm volatility of the USDBRL.

In this work, we focus on two attractive crowd-based business models, i.e., user-generated content creation and freight-matching long-haul trucking. First, as elaborated in CHAPTER 2, we consider a game theoretical modeling approach for understanding the operation of non-profit UGC platforms that rely on users to create content and maintain financial sustainability. In particular, we examine several interesting research questions with practical importance and unique contributions to the literature. These research questions mainly investigate how changes in critical business factors influence the platforms' strategic effort allocation, user participation, and overall performance. Second, in CHAPTER 3, we focus on the flourishing freight-matching businesses that rely on crowdsourced drivers for long-haul trucking. In particular, although the practice suggests that shippers' ordering behaviors of freight-matching services may remarkably impact crowdsourced drivers' bidding behaviors, the literature has yet to examine this issue formally. Therefore, we collect industrial data and construct a strict empirical schema for understanding the association between shippers' order timing and freight-matching performance. Besides, by deliberately building a theoretical modeling framework and using a data-driven estimation of model parameters, we are able to simulate the freight-matching performance of adopting our empirical findings and evaluate the practical value of our study. By investigating these two prominent business models, we aim to understand the advantages of crowdsourcing businesses and the role of crowds in nowadays' business innovations. Besides, we also provide valuable managerial insights for business runners who are interested in this "young" market of crowdsourcing businesses. / Business Administration/Strategic Management

[pt] Tradicionalmente, muitos trabalhos têm estudado os efeitos das intervenções cambiais esterilizadas nas taxas de câmbio, tanto empiricamente quanto teoricamente, encontrando resultados mistos. Mais recentemente, a literatura de finanças internacionais têm procurado explicar os desvios na Paridade Coberta da Taxa de Juros (PCJ), que vem sendo observado entre as moedas das economias desenvolvidas após a Grande Crise Financeira de 2008. Neste trabalho, ligamos as duas literaturas ao estudar o efeito das intervenções cambiais nos desvios na paridade coberta de juros. Nossa amostra consiste nas intervenções realizadas pelo Banco Central do Brasil entre os anos de 2009 e 2020. Este período contempla o programa de intervenções pré-anunciadas de 2013, implementado no contexto do Taper Tantrum, e que já mostrou ter afetado significantemente as taxas de câmbio (Chamon, Garcia e Souza (2017) ). Para avaliar os efeitos, construímos uma série contrafactual utilizando a metodologia ArCo, desenvolvida por Carvalho, Masini e Medeiros (2018), e também estimando funções impulso resposta utilizando Local Projection, desenvolvida por Jordà (2005). Os resultados indicam que a venda de dólares no mercado futuro aumentam os desvios na PCJ, enquanto que compras de dólares tem o efeito oposto. A oferta de dólares via contratos de recompra diminui os desvios no curto prazo. As intervenções no mercado a vista apresentam resultados inconclusivos. / [en] Traditionally, much has been written about the effects of FX (foreign exchange) sterilized interventions on exchange rates, both theoretically and empirically, with mixed results. More recently, the international finance literature has tried to explain the deviations from the well-known Covered Interest Parity (CIP) condition that have, since the 2008 Great Financial Crisis, arisen among advanced economies currencies. Here, we originally merge these two strands of the literature by analyzing the effects of sterilized FX interventions on the CIP (Covered Interest Parity) deviation. Our sample is composed of Brazilian Central Bank FX interventions between 2009 and 2020. This period contains a major program of announced FX interventions in response to the Taper Tantrum, in 2013, which has already been shown to have significantly affected the level of the exchange rate (Chamon, Garcia, and Souza (2017)). To gauge the effects, we build a counterfactual employing the ArCo methodology, developed by Carvalho, Masini, and Medeiros (2018), and also make use of Jordà (2005) Local Projections. The results indicate that selling US dollars in the futures market increases CIP deviations while buying US dollar futures has the opposite effect. Offering US dollar repo credit lines points to a short-lived decrease in the deviation. The number of sterilized sales or purchases of spot currency seems not to be high enough to lead to conclusive results.

Este trabalho busca trazer robustez aos resultados observados em estudos comparativos entre escolas públicas e privadas do Brasil, que indicam uma maior capacidade da rede particular de ensino em gerar qualidade educacional. Para isso, uma abordagem quantílica, baseada na seleção em observadas, foi realizada. Vale ressaltar que, ao contrário de outras abordagens, a realizada nesta dissertação tem inferência assintótica. A base de dados utilizada foi a do SAEB de 2005, para as provas de matemática de oitava série. Novamente foi evidenciada uma superioridade das escolas privadas, mesmo controlando para diversas covariadas de alunos, professores e escolas. Este fato fortalece a possibilidade de implantação de políticas de cupons para escolas particulares, apesar de haver a necessidade de estudos adicionais sobre o assunto. Em relação às covariadas que reduziriam a distância entre as distribuições de notas de alunos de escolas públicas e privadas, constatouse que fatores determinantes do grupo de alunos na escola e na sala (peer group effects) seriam os mais importantes. Isso corrobora com resultados de outros trabalhos que evidenciam a importância desses fatores para explicar a maior efetividade das escolas privadas em relação às escolas públicas. / This paper aims at bringing strength to the results observed in other studies that point out a larger ability of the private school network to generate quality education in Brazil. To achieve that result, this study applies a quantile approach based on the selection on observable variables. Note that unlike other approaches , the one applied in this dissertation has asymptotic inference. The data base used in this study was that of SAEB 2005 for math tests in the 8th grade. As in other studies, here again the superiority of the private schools was made evident, even though various students\', teachers\' and schools\' covariates are controlled. This result strengthens the possibility of a policy of quotas for private schools, although additional studies on the subject are necessary. In relation to the variables that would reduce the distance between the grades distributions of students on public and private schools, peer group effects were observed to be the more important ones. These results are similar to the ones observed in other studies that point out the importance of the peer group effects to explain the higher effectiveness of the private schools in comparison to the public schools.

In this dissertation, I propose a reductive account of causation. This account may be stated as follows: Causation: c is a cause of e within a possibility horizon ℓ if a) c is process-connected to e, and b) e security-depends on c within ℓ. More precisely, my suggestion is that there are two kinds of causal relata: instantaneous events (defined in Chapter 4) and possibility horizons (defined in Chapter 5). Causation is a ternary relation between two actual instantaneous events - the cause c and the effect e - and a possibility horizon ℓ. I argue that causation has a dual nature: on the one hand, a cause must be connected to its effect via a genuine process; on the other hand, a cause must make a difference to its effect. The first condition - namely, the condition of process-connection (defined in Chapter 6) - captures the sense in which a cause must be connected to its effect via a genuine process. This condition allows my account to separate causation from mere correlation, distinguish genuine causes from preempted backups, and capture how a cause must be at the right level of detail relative to its effect (Chapter 7). The second condition - namely, the condition of security-dependence (defined in Chapter 8) - captures the sense in which a cause must make a difference to its effect. This condition allows my account to yield intuitively correct verdicts on the counterexamples to the transitivity and intrinsicness of causation, resolve the problem of profligate omissions, accommodate structurally isomorphic but causally different cases, and handle contrastive causal claims (Chapter 9 and 10). Finally, my proposed account of causation logically entails restricted versions of three important principles of causal reasoning concerning the sufficiency of counterfactual dependence for causation, and the transitivity and intrinsicness of causation (Chapter 11).