Center for Operations Research and Econometrics


Center for Operations Research and Econometrics

Time filter
Source Type

Bogolubsky L.,Moscow State University | Gusev G.,Moscow Institute of Physics and Technology | Raigorodskii A.,Moscow Institute of Physics and Technology | Tikhonov A.,Yandex | And 4 more authors.
Advances in Neural Information Processing Systems | Year: 2016

In this paper, we consider a non-convex loss-minimization problem of learning Supervised PageRank models, which can account for features of nodes and edges. We propose gradient-based and random gradient-free methods to solve this problem. Our algorithms are based on the concept of an inexact oracle and unlike the state-of-the-art gradient-based method we manage to provide theoretically the convergence rate guarantees for both of them. Finally, we compare the performance of the proposed optimization methods with the state of the art applied to a ranking task. © 2016 NIPS Foundation - All Rights Reserved.

Farina P.,Diego Portales University | San Martin E.,University of Santiago de Chile | San Martin E.,Center for Operations Research and Econometrics | Preiss D.D.,University of Santiago de Chile
Computers and Education | Year: 2015

This work studies the relation between computer use for reading activities and academic literacy in 15-year-old students in Chile, Uruguay, Spain, and Portugal. Data used is from the PISA 2009 test. Special attention is given to potential bias problems when the computer use is an endogenous variable. Few studies in this area address this issue: existing literature has shown that different types of computer use have different implications on performance. The limitations of observational data have also been emphasized to establish cause-effect relations between computer use and academic performance. It is important, however, to consider the computer use endogeneity hypothesis (above all at home) since students decide on the frequency of computer use at home. The results found show that by controlling for endogeneity, computer use for reading is not related to reading performance neither in digital or printed format, with the exception of Chile that shows a negative relation in the case of reading from a printed format. The results considering endogeneity differ considerably from results when endogeneity is not taken into account. The work shows the relevance of experimental type studies in order to make sound statements with regard to the computer use and academic performance relation. In turn, school reading activities in a digital environment are suggested that could have an impact on reading performance. © 2014 Elsevier Ltd. All rights reserved.

Abada I.,Engie Inc | Ehrenmann A.,Engie Inc | Smeers Y.,Center for Operations Research and Econometrics
Operations Research | Year: 2017

Since its initial development, long-term contracts have been associated with the gas industry in all regions of the world. This was also the case in Europe where natural gas trade was, for a long time, dominated by bilateral long-term agreements between producers and midstreamers. These contracts fixed a minimum volume to be exchanged (take or pay) and indexed the gas price using a price formula that usually referred to oil product prices. These arrangements allowed market risk sharing between the producer (who takes the price risk) and the midstreamer (who takes the volume risk). They also offered risk hedging since oil is considered as a trusted commodity by investors. The fall of the European natural gas demand, combined with the increase of the oil price, favored the emergence of a gas volume bubble that caused significant losses for most of the European midstreamers bound by long-term agreements. As a result, the downstream part of the industry brought forward the idea of indexing contracts on gas spot prices. In this paper, we present an equilibrium model that endogenously captures the contracting behavior of both producer and midstreamer, who strive to hedge their profit-related risk. Players can choose between gas forward and oil-indexed contracts. Using the model, we show that (i) contracting can reduce the trade risk for both producer and midstreamer; (ii) oil-indexed contracts should be signed only when oil and gas spot prices are well correlated, otherwise, these contracts hold less interest for risk mitigation; (iii) contracts are best suited when the upstream cost structure is mainly driven by capital costs; and (iv) a high level of risk aversion from the midstreamer might deprive upstream investments and downstream consumer surplus. © 2017 INFORMS.

Auray S.,CREST Ensai | Auray S.,University of the Littoral Opal Coast | De Wolf D.,University of the Littoral Opal Coast | Smeers Y.,Center for Operations Research and Econometrics
RAIRO - Operations Research | Year: 2015

In this paper, we formulate and solve a real life coal blending problem using a Column Generation Approach. The objective of the model is to prescribe optimal mixes of coal to produce coke. The problem is formulated as a mixed integer program. It involves various types of constraints arising from technical considerations of the blending process. The model also incorporates nonlinear constraints. It results in a large-scale problem that cannot be solved by classical operations research methods. Defining three heuristic methods based on column generation techniques, this paper proposes reasonable solutions for the industry. © EDP Sciences, ROADEF, SMAI 2014.

Haedo C.,University of Bologna | Mouchart M.,Institute Of Statistique | Mouchart M.,Center for Operations Research and Econometrics
Spatial Statistics | Year: 2015

This paper develops new statistical and computational methods for the automatic detection of spatial clusters displaying an over- or under- relative specialization spatial pattern. A probability model is used to provide a basis for a space partition into clusters representing homogeneous portions of space as far as the probability of locating a primary unit is concerned. A cluster made of contiguous regions is called an agglomeration. A greedy algorithm detects specialized agglomerations through a model selection criteria. A random permutation test evaluates whether the contiguity property is significant. Finally this algorithm is run on Argentinean data. Evaluating the proposed methodology concludes the paper. © 2014 Elsevier Ltd.

Van Vyve M.,Center for Operations Research and Econometrics
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

The fixed-charge transportation problem is a fixed-charge network flow problem on a bipartite graph. This problem appears as a subproblem in many hard transportation problems, and has also strong links with the challenging big-bucket multi-item lot-sizing problem. We provide a polyhedral analysis of the polynomially solvable special case in which the associated bipartite graph is a path. We describe a new class of inequalities that we call "path-modular" inequalities. We give two distinct proofs of their validity. The first one is direct and crucially relies on sub- and super-modularity of an associated set function, thereby providing an interesting link with flow-cover type inequalities. The second proof is by projecting a tight extended formulation, therefore also showing that these inequalities suffice to describe the convex hull of the feasible solutions to this problem. We finally show how to solve the separation problem associated to the path-modular inequalities in O(n3) time. © 2011 Springer-Verlag.

Claro M.,University of Santiago de Chile | Cabello T.,University of Santiago de Chile | San Martin E.,University of Santiago de Chile | San Martin E.,Center for Operations Research and Econometrics | And 2 more authors.
Computers and Education | Year: 2014

This paper provides evidence that helps understand the digital divide in education. It does so by comparing the effect of economic, social and cultural status (ESCS) on the digital skills of Chilean students compared with mathematics and language. This comparison is made using national standardized tests. The marginal effect of a group of variables measuring student ESCS was compared both as a whole and separately using multivariate linear regression analysis. The results show that the marginal effect of ESCS as a whole on students' digital skills was equal to the effect on mathematics and greater than the effect on language. Furthermore, the results show that the parents' level of education was the most relevant factor of ESCS for explaining student performance on the digital test, more so than for mathematics and language. These findings challenge the belief that the Internet would reduce economic, social and cultural inequalities in new generations. Instead, they reveal that the gap among Chilean students tends to perpetuate or widen when comparing performance in mathematics and language with performance in the digital domain. At the same time, by comparing national test results, this paper offers empirical evidence for the existence of a second digital divide in the field of education, a concept which is widely discussed at a theoretical level but with little empirical support to date. © 2014 Elsevier Ltd.

Tancrez J.-S.,Catholic University of Louvain | Chevalier P.,Catholic University of Louvain | Chevalier P.,Center for Operations Research and Econometrics | Semal P.,Catholic University of Louvain
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013

In this paper, we present a bounding methodology that allows to compute a tight bound on the throughput of fork-join queueing networks with blocking and with general service time distributions. No exact models exist for queueing networks with general service time distributions and, consequently, bounds are the only certain information available. The methodology relies on two ideas. First, probability mass fitting (PMF) discretizes the service time distributions so that the evolution of the modified system can be modelled by a discrete Markov chain. Second, we show that the critical path can be computed with the discretized distributions and that the same sequence of jobs offers a bound on the original throughput. The tightness of the bound is shown on computational experiments (error on the order of one percent). Finally, we discuss the extension to split-and-merge networks and the approximate estimations of the throughput. © 2013 Springer-Verlag.

Loading Center for Operations Research and Econometrics collaborators
Loading Center for Operations Research and Econometrics collaborators