Instituto de Matemática de Bahía Blanca

Causal graph extraction from news has the potential to aid in the understanding of complex scenarios. In particular, it can help explain and predict events, as well as conjecture about possible cause-effect connections. However, limited work has addressed the problem of large-scale extraction of causal graphs from news articles. This article presents a novel framework for extracting causal graphs from digital text media. The framework relies on topic-relevant variables representing terms and ongoing events that are selected from a domain under analysis by applying specially developed information retrieval and natural language processing methods. Events are represented as event-phrase embeddings, which make it possible to group similar events into semantically cohesive clusters. A time series of the selected variables is given as input to a causal structure learning techniques to learn a causal graph associated with the topic that is being examined. The complete framework is applied to the New York Times dataset, which covers news for a period of 246 months (roughly 20 years), and is illustrated through a case study. An initial evaluation based on synthetic data is carried out to gain insight into the most effective time-series causality learning techniques. This evaluation comprises a systematic analysis of nine state-of-the-art causal structure learning techniques and two novel ensemble methods derived from the most effective techniques. Subsequently, the complete framework based on the most promising causal structure learning technique is evaluated with domain experts in a real-world scenario through the use of the presented case study. The proposed analysis offers valuable insights into the problems of identifying topic-relevant variables from large volumes of news and learning causal graphs from time series.

Municipal solid waste management is a paramount activity in modern cities due to environmental, social and economic problems that can arise when mishandled. In this work, the sequencing of micro-routes in the Argentine city of Bahía Blanca is addressed, which is modelled as a vehicle routing problem with travel time limit and the vehicle's capacity. Particularly, we propose two mathematical formulations based on mixed integer programming and we solve a set of instances of the city of Bahía Blanca based on real data. Moreover, with this model, we estimate the total distance and travel time of the waste collection and use this data to analyse the possibility of installing a transfer station. The results demonstrate the competitiveness of the approach to resolve realistic instances of the target problem and suggest the convenience of installing a transfer station in the city considering the reduction of the travelled distance.

Abstract The COVID-19 pandemic forced almost all professional and amateur sports to be played without attending crowds. Thus, it induced a large-scale natural experiment on the impact of social pressure on decision making and behavior in sports fields. Using a data set of 1027 rugby union matches from 11 tournaments in 10 countries, we find that home teams have won less matches and their point difference decreased during the pandemic, shedding light on the impact of crowd attendance on the home advantage of sports teams.

Successful modeling and prediction depend on effective methods for the extraction of domain-relevant variables. This paper proposes a methodology for identifying domain-specific terms. The proposed methodology relies on a collection of documents labeled as relevant or irrelevant to the domain under analysis. Based on the labeled document collection, we propose a supervised technique that weights terms based on their descriptive and discriminating power. Finally, the descriptive and discriminating values are combined into a general measure that, through the use of an adjustable parameter, allows to independently favor different aspects of retrieval such as maximizing precision or recall, or achieving a balance between both of them. The proposed technique is applied to the economic domain and is empirically evaluated through a human-subject experiment involving experts and non-experts in Economy. It is also evaluated as a term-weighting technique for query-term selection showing promising results. We finally illustrate the applicability of the proposed technique to address diverse problems such as building prediction models, supporting knowledge modeling, and achieving total recall.

The design of the bus network is a complex problem in modern cities, since different conflicting objectives have to be considered, from both the perspective of bus companies and the citizens. This article presents a multiobjective model for designing a sustainable public transportation network that simultaneously optimizes the covered travel demands by passengers, the total travel time, and the generated pollution. The proposed model is solved using exact weighted sum and a heuristic procedure based on the standard shortest path problem. Preliminary tests were performed in small real-world instances of Montevideo, Uruguay. Experiments allowed obtaining a set of compromising solutions that in turn allow exploring different trade-off among the optimization criteria. The proposed heuristic was competitive, being able to find a good compromising solution in short computing times.

Although the identification of topics and sentiments from social media content has attracted substantial research, little work has been carried out on the extraction of causal relationships among those topics and sentiments. This article proposes a methodology aimed at building a causal graph where nodes represent topics and emotions extracted from social media users’ posts. To illustrate the proposed methodology, we collected a large multi-year dataset of tweets related to different editions of the G20 summit, which was locally indexed for further analysis. Topic-relevant queries are crafted from phrases extracted by experts from G20 output documents on four main recurring topics, namely government, society, environment and health and economics. Subsequently, sentiments are identified on the retrieved tweets using a lexicon based on Plutchik’s wheel of emotions. Finally, a causality test that uses stochastic dominance is applied to build a causal graph among topics and emotions by exploiting the asymmetries of explaining a variable from other variables. The applied causality discovery process relies on observational data only and does not require any assumptions of linearity, parametric definitions or temporal precedence. In our analysis, we observe that although the time series of topics and emotions always show high correlation coefficients, stochastic causality provides a means to tell apart causal relationships from other forms of associations. The proposed methodology can be applied to better understand social behaviour on social media, offering support to decision and policy making and their communication by government leaders.

INTRODUCTION: Financial protection from healthcare spending has become an important objective to be addressed by health systems all over the world. A common strategy used to assess financial protection from health care is to estimate the proportion of the population for which out-of-pocket expenditures made at the moment of receiving health services (OOP) might affect the consumption of other goods and services. To this end, two groups of indicators have been developed: catastrophic health expenditure (CHE) and impoverishing health expenditure (IHE). This work aims to investigate how CHE and IHE evolved in Argentina and how equitable was distributed between 1985 and 2018. METHODOLOGY: we estimated CHE and IHE measures, concentration indexes and concentration curves for all studied periods. In addition, we performed dominance analysis of concentration curves in order to assess changes in the distribution of CHE. RESULTS: In 2017/18, 9.57 % of Argentina’s population incurred in CHE using a 10 % of total expenditure (EXP) threshold, 5.81 % using a 15 % of EXP, 4.52 % using a 25 % of EXP net of food spending (ATP), and 1.87 % using a 40 % of ATP. All CHE headcount measures dropped considerably between 1996/97 and 2017/18. IHE measures resulted in nearly zero values. The distribution of CHE was found to be progressive in all periods applying different thresholds. Dominance analysis and CI show that 2004/05 was the most progressive period. However, dominance between curves was only found using low specificity criteria. DISCUSSION: We found evidence of higher financial protection in the most recent studied period and progressivity of CHE in all periods. A further question to be assessed is whether the lower CHE and progressivity in its distribution is a consequence of an effective public policy or difficulties to access health care.

Abstract From a panel of administrative data of contributors to the Argentine pension system, a survival analysis was carried out to compare the time in months until contributor status is lost for individuals who reach the minimum retirement age. The analysis involves two periods: one where the fully funded scheme was in force and the other with the pure pay‐as‐you‐go scheme in effect. It is observed that the survival curve of the contributor status during the second period is always above that in the first one, indicating that the probability of maintaining this status was always higher during the second period. Likewise, it was found that belonging to the cohort/group of the first period is significant in explaining the risk of losing contributor status, even controlling for other relevant variables. This result contrasts with the predictions of the World Bank about the best incentives that the fully funded scheme would generate for the supply of formal labor. Related Articles Angelaki, Marina, and Leandro N. Carrera. 2015. “Radical Pension Reforms after the Crisis: A Comparative Analysis of Argentina and Greece.” Politics & Policy 43(3): 378–400. https://doi.org/10.1111/polp.12117 . Hazakis, Konstantinos J. 2015. “The Political Economy of Economic Adjustment Programs in the Eurozone: A Detailed Policy Analysis.” Politics & Policy 43(6): 822–54. https://doi.org/10.1111/polp.12141 . Jakee, Keith, and Keith Stacy. 2021. “Revisiting Pension Reform in Sweden: The Role of Corporatism (and Why it Matters).” Politics & Policy 49(3): 651–81. https://doi.org/10.1111/polp.12405 .

We experimentally approach the discursive dilemma to gain insight into people's procedural appropriateness judgments. We relied on a vignette in which three people had formed opinions about two skills (premises) of a candidate to decide whether to hire her/him (conclusion). The dilemma arises when different outcomes (hire vs. not hire) are achieved depending on whether the majority opinion is independently considered for each premise or for the global conclusion of each judge. Participants were asked to choose the procedure they thought to be more appropriate to reach a decision. In Experiment 1, we found a leniency effect (a bias to prefer the aggregation procedure that led to hiring the candidate), which was reduced by introducing the participant as a juror with an exogenously provided negative opinion about the candidate's skills. In Experiment 2, we replicated the opinion effect, even when subjects did not participate as jury members. In Experiment 3, we found that the leniency bias was only reduced when participants' negative opinion was aligned with a majority of negative premises, but not with a majority of negative conclusions. We discuss present findings in terms of the identification of empirical regularities that may affect people's procedural legitimacy judgments.

El objetivo de este trabajo es evaluar la dinámica semanal de precios del sector de galletitas en el marco de una economía con inflación moderada, tal como lo es la argentina. Empleando datos de frecuencia semanal, se estiman cinco versiones de la función que relaciona a la variación semanal de los precios de las galletitas con sus posibles determinantes. Cuatro de esas estimaciones han sido obtenidas mediante mínimos cuadrados generalizados (MCG), en tanto que la restante corresponde a una especificación con efectos aleatorios (EA). Se identificaron posibles influencias del nivel de concentración de la oferta, las características de cada producto (tamaño, sabor, tipo), la situación del mercado cambiario, ciertos efectos temporales y las variaciones en los precios de los insumos sobre la dinámica de precios de las galletitas. Sin embargo, solamente los efectos de las variaciones en los precios de los insumos se presentaron robustos. El signo de estos efectos robustos coincide siempre con el esperado, a excepción del caso de la harina de trigo.

In this paper we describe a framework to uncover potential causal relations between event mentions from streaming text of news media. This framework relies on a dataset of manually labeled events to train a recurrent neural network for event detection. It then creates a time series of event clusters, where clusters are based on BERT contextual word embedding representations of the identified events. Using these time series dataset, we assess four methods based on Granger causality for inferring causal relations. Granger causality is a statistical concept of causality that is based on forecasting. It states that a cause occurs before the effect, and the cause produces unique changes in the effect, so past values of the cause help predict future values of the effect. The four analyzed methods are the pairwise Granger test, VAR(1), BigVar and SiMoNe. The framework is applied to the New York Times dataset, which covers news for a period of 246 months. This preliminary analysis delivers important insights into the nature of each method, identifies differences and commonalities, and points out some of their strengths and weaknesses.

Damages for personal injuries is a field prone to gender biases. Empirical exploration of the issue, however, is far from being simple, especially in Civil Law countries, given a pervasive lack of transparency and explicitness on the details of reasoning and treatment of numerical values. Accounting for that feature, our research sketches a canon of analysis that goes beyond the words. To deal with risks of cherry picking and inaccurate sample design, our database encompasses all the decisions made during the span time selected (more than 20,000) by an Appellate Court. Scrapers designed ad hoc have been instrumental to that aim. In our case study, the well-known earnings gap, usually assumed a cause of other, second order gaps, ought not to be mirrored in magnitude by pecuniary damages, because the “shadow price” of non-remunerated activities must be computed together with earnings lost, what should counterbalance the latter derivative gap. Nonpecuniary damages, in turn, must be independent of earnings, being theoretically free from that effect. Nonetheless, we found differences with statistical significance in any of them, in favor of men. Some reverse engineering in search of the primary source of the gap leads to find a systematic bias in percentages of disability against women, even in cases where the predictable result should be the opposite. In more general terms, the said obscurity on the treatment of numerical values, usually covered by rhetoric, renders difficult any honest scrutiny of systematic biases on the matter without the assistance of high technology and some sophistication, and shields decisions to criticism. Accordingly, it helps perpetuate gaps whenever existed.

The study of the interactions among intentional agents, with rationality being the main source of intentional behavior, requires mathematical tools capable of capturing systemic effects. Here, we choose an alternative toolbox based on Category Theory. We examine potential level-agnostic formalisms, presenting three categories: PR, G, and an encompassing one, I. The latter allows for representing dynamic rearrangements of the interactions among different agents. Systems represented in I capture the dynamic interactions among the interfaces of their sub-agents, changing the connections among them based on their internal states. We illustrate the expressive power of this formalism in four different instances, providing practitioners with a toolbox for representing cases of interest and facilitating their modular analysis.

One of the consequences of persistent technological change is that it force individuals to make decisions under extreme uncertainty. This means that traditional decision-making frameworks cannot be applied. To address this issue we introduce a variant of Case-Based Decision Theory, in which the solution to a problem obtains in terms of the distance to previous problems. We formalize this by defining a space based on an orthogonal basis of features of problems. We show how this framework evolves upon the acquisition of new information, namely features or values of them arising in new problems. We discuss how this can be useful to evaluate decisions based on not yet existing data.

Let $μ$ be a non-negative Borel measure on $R^d$ satisfying that the measure of a cube in $R^d$ is smaller than the length of its side raised to the $n$-th power, $0

Causal graph extraction from news has the potential to aid in the understanding of complex scenarios.In particular, it can help explain and predict events, as well as conjecture about possible cause-effect connections.However, limited work has addressed the problem of large-scale extraction of causal graphs from news articles.This article presents a novel framework for extracting causal graphs from digital text media.The framework relies on topic-relevant variables representing terms and ongoing events that are selected from a domain under analysis by applying specially developed information retrieval and natural language processing methods.Events are represented as eventphrase embeddings, which make it possible to group similar events into semantically cohesive clusters.A time series of the selected variables is given as input to a causal structure learning techniques to learn a causal graph associated with the topic that is being examined.The complete framework is applied to the New York Times dataset, which covers news for a period of 246 months (roughly 20 years), and is illustrated through a case study.An initial evaluation based on synthetic data is carried out to gain insight into the most effective time-series causality learning techniques.This evaluation comprises a systematic analysis of nine state-of-the-art causal structure learning techniques and two novel ensemble methods derived from the most effective techniques.Subsequently, the complete framework based on the most promising causal structure learning technique is evaluated with domain experts in a real-world scenario through the use of the presented case study.The proposed analysis offers valuable insights into the problems of identifying topic-relevant variables from large volumes of news and learning causal graphs from time series.

ABSTRACT Contribution density is not the same as labor formality; it refers to the amount or proportion of contributions per worker that can be accumulated on average in a given period. The objective of this article is to identify the determinants of contribution density to the Argentine pension system from administrative data from 1996 to 2021. An econometric model is proposed to explain the contribution density in eight cohorts based on sociodemographic factors and job characteristics, using the negative binomial regression method. It is found that the density of contributions is low and depends on the branch of economic activity, the jurisdiction where the company is located, the individual's income and gender, and the age of enrollment. The results are similar to those found by other works that study contribution density in Argentina and Latin American countries.

Abstract In this paper endpoint entropy Fefferman–Stein bounds for Calderón–Zygmund operators introduced by Rahm (J Math Anal Appl 504(1):Paper No. 125372, 2021) are extended to iterated Coifman–Rochberg–Weiss commutators.

We present a fuzzy version of the Group Identification Problem ("Who is a J?") introduced by Kasher and Rubinstein (1997). We consider a class $N = \{1,2,\ldots,n\}$ of agents, each one with an opinion about the membership to a group J of the members of the society, consisting in a function $π: N \to [0; 1]$, indicating for each agent, including herself, the degree of membership to J. We consider the problem of aggregating those functions, satisfying different sets of axioms and characterizing different aggregators. While some results are analogous to those of the originally crisp model, the fuzzy version is able to overcome some of the main impossibility results of Kasher and Rubinstein.

Kaushik Basu presents the Traveler’s Dilemma (TD) as a challenge to Game Theory. This challenge has been experimentally investigated. When faced with Basu’s version of the TD, participants (including experts in game theory) behave inthe way Basu suggests. However, a little change in the game turns out to reverse participants’ choices. The question is, then, whether it is possible to provide an account of the main empirical findings as consequences of rational choices (i.e., torationalize them). There are several proposals in the literature but none of them provides a satisfactory account for why experts in game theory playing against each other usually reject the only undominated option of the TD. The goal of this article is to suggest an alternative proposal that fixes this problem.