BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Statistics and Operations Research - ECPv5.1.6//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Department of Statistics and Operations Research
X-ORIGINAL-URL:https://stat-or.unc.edu
X-WR-CALDESC:Events for Department of Statistics and Operations Research
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20160313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20161106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20170312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20171105T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160120T143000
DTEND;TZID=America/New_York:20160120T153000
DTSTAMP:20210515T201830
CREATED:20170216T012016Z
LAST-MODIFIED:20170216T012024Z
UID:2748-1453300200-1453303800@stat-or.unc.edu
SUMMARY:STOR Colloquium: Noah Forman\, University of Oxford
DESCRIPTION:Exchangeability and continuum random trees \nA sequence of random variables is exchangeable if its distribution is invariant under finite permutations of indices. De Finetti showed that such sequences may be viewed as a mixtures of i.i.d. (independent\, identically distributed) sequences. The Chinese restaurant process (CRP) is a simple model related to many exchangeable objects\, which has recently gained prominence in connection with the clustering problem in machine learning. Continuum random trees (CRTs)\, on the other hand\, are related to various models (branching\, fragmentation\, coalescence) that have arisen from population genetics. This talk discusses connections between CRPs and CRTs. We focus on ongoing research concerning a Chinese restaurant with reseating that arises within a randomly evolving tree\, and the continuum limits of these processes\, with far-flung connections to Lévy process local times\, excursion theory\, and Wright-Fisher diffusions.
URL:https://stat-or.unc.edu/event/stor-colloquium-noah-forman-university-of-oxford/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160127T143000
DTEND;TZID=America/New_York:20160127T153000
DTSTAMP:20210515T201830
CREATED:20170216T012123Z
LAST-MODIFIED:20170216T012123Z
UID:2750-1453905000-1453908600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Sanchayan Sen\, Eindhoven University of Technology
DESCRIPTION:Asymptotics of high dimensional random structures: Probabilistic combinatorial optimization\, stochastic geometry and random matrices \nOne key focus of probability theory over the last few years has been on understanding asymptotics in the context of high dimensional structures. The aim of this talk is to give an overview of three major themes that have arisen in my work. More precisely: \na) Probabilistic combinatorial optimization: One major conjecture in probabilistic combinatorics\, formulated by statistical physicists using non-rigorous arguments and enormous simulations in the early 2000s\, is as follows: for a wide array of random graph models on n vertices and degree exponent tau>3\, typical distance both within maximal components in the critical regime as well as on the minimal spanning tree on the giant component in the supercritical regime scale like n^{(min(tau\,4) -3)/(min(tau\,4) -1)}. In other words\, the degree exponent determines the universality class the network belongs to. The mathematical machinery available at the time was insufficient for providing a rigorous justification of this conjecture. We report on recent progress in proving this conjecture and characterizing these universality classes in a broader sense. \nb) Stochastic geometry: In the case of spatial systems\, less precise results are known. We discuss a recent result concerning the length of spatial minimal spanning trees that answers a question raised by Kesten and Lee in the 90’s\, the proof of which relies on a variation of Stein’s method and a quantification of a classical argument in percolation theory. \nc) Random matrix theory: Many random matrix ensembles arise naturally in statistics and physics. In the final part of the talk\, we discuss properties of empirical spectral distributions of two such ensembles.
URL:https://stat-or.unc.edu/event/stor-colloquium-sanchayan-sen-eindhoven-university-of-technology/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160129T143000
DTEND;TZID=America/New_York:20160129T153000
DTSTAMP:20210515T201830
CREATED:20170216T012237Z
LAST-MODIFIED:20170216T012237Z
UID:2752-1454077800-1454081400@stat-or.unc.edu
SUMMARY:STOR Colloquium: Nicolas Fraiman\, Harvard University
DESCRIPTION:Effects of limited choice and cooperation in network formation\, epidemics and evolution. \nHow does having structural or dynamical constraints affect the outcomes of stochastic models? In this talk I will discuss changes obtained from the addition of limited choice\, cooperation or competition into processes that describe the evolution of networks and populations. I will analyze three models motivated by applications in communication networks\, spread of epidemics\, and evolutionary biology. The focus is on asymptotic behavior and phase transitions. \nBased on joint works with Luc Devroye\, Gabor Lugosi and Martin Nowak.
URL:https://stat-or.unc.edu/event/stor-colloquium-nicolas-fraiman-harvard-university/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160201T143000
DTEND;TZID=America/New_York:20160201T153000
DTSTAMP:20210515T201830
CREATED:20170216T012340Z
LAST-MODIFIED:20170216T012340Z
UID:2754-1454337000-1454340600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Sayan Banerjee\, University of Warwick
DESCRIPTION:Couplings and geometry \nA coupling of (the laws of) two Markov processes specifies a particular construction of copies of the two processes simultaneously on the same space. They have a long history and find numerous applications in probability and analysis\, ranging from yielding bounds on the mixing times of Markov chains to studying harmonic maps. \nIt is natural to ask whether we can construct a coupling where the coupled processes actually meet (successful coupling). If such a coupling exists\, how fast can we make them meet (coupling rate)? It turns out that this question has deep connections with the generator of the Markov process and the geometry of the underlying space. \nIn this talk\, I will give an overview of some results in this area. In particular\, we will focus on general elliptic diffusions on Riemannian manifolds\, and show how geometry (dimension of the isometry group\, flows of isometries\, Killing vector fields and dilation vector fields) plays a fundamental role in relating the space and the generator of the diffusion to the coupling rate. I will also briefly describe efficient coupling techniques for some nilpotent diffusions using ideas from the theory of infinite dimensional Brownian motion. \nThis is joint work with W.S. Kendall.
URL:https://stat-or.unc.edu/event/stor-colloquium-sayan-banerjee-university-of-warwick/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160203T143000
DTEND;TZID=America/New_York:20160203T153000
DTSTAMP:20210515T201830
CREATED:20170216T012455Z
LAST-MODIFIED:20170216T012455Z
UID:2756-1454509800-1454513400@stat-or.unc.edu
SUMMARY:STOR Colloquium: Yu-Ting Chen\, Harvard University
DESCRIPTION:Stochastic interacting systems on graphs \nFor more realistic modeling\, there have been significant interests for the use of general graphs for interacting particle systems arising from biological or social contexts. However\, the generality of spatial structure can lead to fundamental issues. They include the missing link to the stochastic PDE method that can give very detailed and clean information of interacting particle systems after rescaling\, and the question of which graph parameters are essential to describe certain probability laws of interest.\nIn this talk\, I will discuss voter models and related methods\, with an emphasis on the context of general graphs. I will demonstrate recent progress for some questions from Aldous for voter models and the so-called benefit-to-cost ratios first discovered by Ohtsuki\, Hauert\, Lieberman\, and Nowak for evolutionary games that are variants of voter models. In particular\, I will explain some diffusion approximation results for voter models on general graphs\, which may provide new insights for related interacting particle systems on large graphs.
URL:https://stat-or.unc.edu/event/stor-colloquium-yu-ting-chen-harvard-university/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160208T143000
DTEND;TZID=America/New_York:20160208T153000
DTSTAMP:20210515T201830
CREATED:20170216T012546Z
LAST-MODIFIED:20170216T012546Z
UID:2758-1454941800-1454945400@stat-or.unc.edu
SUMMARY:STOR Colloquium: Subhro Ghosh\, Princeton University
DESCRIPTION:Rigidity phenomena in random point sets \nIn several naturally occurring (infinite) random point processes\, we establish that the number of the points inside a bounded domain can be determined\, almost surely\, by the point configuration outside the domain. This includes key examples coming from random matrices and random polynomials. We further explore other random processes where such ”rigidity” extends to a number of moments of the mass distribution. The talk will focus on particle systems with such curious “rigidity” phenomena\, and their implications. We will also talk about applications to natural questions in stochastic geometry and harmonic analysis.
URL:https://stat-or.unc.edu/event/stor-colloquium-subhro-ghosh-princeton-university/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160307T143000
DTEND;TZID=America/New_York:20160307T153000
DTSTAMP:20210515T201830
CREATED:20170216T012651Z
LAST-MODIFIED:20170216T012651Z
UID:2760-1457361000-1457364600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Maria Mayorga\, North Carolina State University
DESCRIPTION:A Nested-Compliance Table Policy for Emergency Medical Service Systems under Relocation \nThe goal of Emergency Medical Service (EMS) systems is to provide rapid response to emergency calls in order to save lives. This paper proposes a relocation strategy to improve the performance of EMS systems. In practice\, EMS systems often use a compliance table to relocate ambulances. A compliance table specifies ambulance base stations as a function of the state of the system. We consider a nested-compliance table\, which restricts the number of relocations that can occur simultaneously. We formulate the nested-compliance table model as an integer programming model in order to maximize expected coverage. We determine an optimal nested-compliance table policy using steady state probabilities of a Markov chain model with relocation as input parameters. These parameter approximations are independent of the exact compliance table used. We assume that there is a single type of medical unit\, single call priority\, and no patient queue. We validate the model by applying the nested-compliance table policies in a simulated system using a real-world data. The numerical results show the benefit of our model over a static policy based on the adjusted maximum expected covering location problem (AMEXCLP). Extensions to include districting decisions and formulate the problem as a stochastic program are also presented.
URL:https://stat-or.unc.edu/event/stor-colloquium-maria-mayorga-north-carolina-state-university/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160411T153000
DTEND;TZID=America/New_York:20160411T163000
DTSTAMP:20210515T201830
CREATED:20170216T012752Z
LAST-MODIFIED:20170216T012752Z
UID:2762-1460388600-1460392200@stat-or.unc.edu
SUMMARY:STOR Colloquium: Lexin Li\, University of California at Berkeley
DESCRIPTION:Tensor Regression and Applications in Neuroimaging Analysis \nClassical regression methods treat covariates (response or predictor) as a vector and estimate a corresponding vector of regression coefficients. Modern applications in medical imaging generate covariates of more complex form such as multidimensional array (tensor). Traditional statistical and computational methods are proving insufficient for analysis of such data due to their ultrahigh dimensionality as well as complex structure. In this talk\, we propose a new family of tensor regression models that reduce the ultrahigh dimensionality to a manageable level\, which in turn leading to efficient model estimation and prediction. Some applications to real neuroimaging data analysis will be discussed
URL:https://stat-or.unc.edu/event/stor-colloquium-lexin-li-university-of-california-at-berkeley/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160418T153000
DTEND;TZID=America/New_York:20160418T163000
DTSTAMP:20210515T201830
CREATED:20170216T011906Z
LAST-MODIFIED:20170216T011906Z
UID:2746-1460993400-1460997000@stat-or.unc.edu
SUMMARY:STOR Colloquium: David Dunson\, Duke University
DESCRIPTION:Probabilistic modeling of big table and networks \nIn many applications\, data consist of high-dimensional complex and highly-structured discrete data. Our focus here is on high-dimensional unordered categorical data\, which arise in epidemiology\, social surveys and brain connectomics. In the first part of the talk\, I will focus on data that can be structured as a multiway contingency table but that otherwise have no obvious structure a priori. For such problems\, we rely on probabilistic tensor factorizations\, introducing new classes of factorizations\, discussing relationships with sparse log-linear models\, sketching theory on rates of convergence\, and considering applications in social science surveys and genomics. In the second part of the talk\, I focus on the case in which the categorical data consist of indicators of connections between pairs of nodes in a network\, motivated in particular by brain connectomic studies. The probability distribution for such network-valued random variables can be conveniently represented via a hierarchical latent space representation. We propose a Bayesian approach to inference and show exciting results in performing inferences on differences in brain structure with phenotypes.
URL:https://stat-or.unc.edu/event/stor-colloquium-david-dunson-duke-university/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160912T153000
DTEND;TZID=America/New_York:20160912T163000
DTSTAMP:20210515T201830
CREATED:20160902T190454Z
LAST-MODIFIED:20170215T212350Z
UID:1860-1473694200-1473697800@stat-or.unc.edu
SUMMARY:STOR Colloquium: Alfredo Garcia\, ARO and University of Florida
DESCRIPTION:Alfredo Garcia\, Army Research Office (ARO) and University of Florida \n \nIterative Mechanisms for Electricity Markets \n \nAbstract: We consider the problem of designing the rules by which dispatch and payment to electricity market participants are gradually adjusted while taking into account network and reliability constraints so as to ensure the market clears with an efficient outcome. Small adjustments (which require minimal information from market participants at each iteration) facilitate the identification of incentives for ensuring truthful reporting of private information. We propose a class of iterative mechanisms and show this class exhibits many desirable properties: incentive compatibility\, efficiency\, individual rationality and (weak) budget balance. We also analyze an iterative mechanism for stochastic market clearing\, a pressing need given the increasing penetration of highly intermittent renewable generation technologies. In this case\, the marginal cost of adjustments may only be estimated with some error. We show that truthful reporting is a Nash equilibrium and the resulting dispatch converges almost surely to the efficient dispatch. \nBased on the paper posted on the Arxiv: http://arxiv.org/abs/1608.08987 \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20160919T153000
DTEND;TZID=America/New_York:20160919T163000
DTSTAMP:20210515T201830
CREATED:20160902T190558Z
LAST-MODIFIED:20170215T212350Z
UID:1862-1474299000-1474302600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Bin Hu\, University of North Carolina at Chapel Hill
DESCRIPTION:Bin Hu \nUniversity of North Carolina at Chapel Hill\, Kenan-Flagler Business School\n \n\n\nIs Reshoring Better than Offshoring? The Effect of Offshore \nSupply Dependence\n \nIn this paper we investigate the effect of offshore supply dependence (OSD) on offshoring-reshoring profit comparisons. We find that OSD hampers a reshoring manufacturer’s responsiveness to demand information updates and may significantly affect offshoring-reshoring comparisons\, such that reshoring may yield lower profits than offshoring in many cases\, including when offshoring has no baseline-cost advantage. We then show that OSD also affects how salient costs such as customs duties and shipping costs influence offshoring-reshoring profit comparisons. We further identify common-component designs as a mitigating measure to make reshoring more appealing under OSD\, and numerically confirm the robustness of our results. \n \nhttp://papers.ssrn.com/sol3/papers.cfm?abstract_id=2645328 \n\n \n \n Refreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-2/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161003T153000
DTEND;TZID=America/New_York:20161003T163000
DTSTAMP:20210515T201830
CREATED:20160906T144800Z
LAST-MODIFIED:20170215T212350Z
UID:1880-1475508600-1475512200@stat-or.unc.edu
SUMMARY:STOR colloquium: Nelson Antunes\, CEMAT/University of Lisbon
DESCRIPTION:Nelson Antunes\nCEMAT/University of Lisbon and University of Algarve \n \nSampling Internet traffic: estimation of flow distributions\nIn this talk\, the focus is on recovering the flow (sequence of packets sharing common attributes) size and duration distributions from sampled Internet traffic using a number of previously introduced packet sampling algorithms. Assuming a basic probabilistic flow model\, the distribution of flow durations is expressed in terms of the distributions of flow sizes and flow IATs (interarrival times between packets). The available estimation methods of the flow size distribution through inversion are then discussed. Inverse equations that allow recovering the distribution of flow IATs from sampled flows and hence the flow duration distribution are presented. An asymptotic approach is also considered to estimate directly the distribution tails of flow durations and sizes from the sampled flow quantities. Both the inversion and asymptotic approaches are evaluated on two Internet traces\, along with a discussion on combining the inversion and asymptotic approaches for a more accurate estimation of both the main body and the tail of the flow distributions. \n \n(This is a joint work with Vladas Pipiras at UNC-Chapel Hill.) \n\n \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-nelson-antunes-cematuniversity-of-lisbon/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161010T153000
DTEND;TZID=America/New_York:20161010T163000
DTSTAMP:20210515T201830
CREATED:20160906T144918Z
LAST-MODIFIED:20170215T212350Z
UID:1882-1476113400-1476117000@stat-or.unc.edu
SUMMARY:STOR colloquium: Joe Guinness\, North Carolina State University
DESCRIPTION:Joe Guinness \nNorth Carolina State University \n \nPermutation methods for sharpening Gaussian process approximations \n \nVecchia’s approximate likelihood for Gaussian process parameters depends on how the observations are ordered\, which can be viewed as a deficiency because the exact likelihood is permutation-invariant. I take the alternative standpoint that the ordering of the observations is an aspect that can be tuned to sharpen the approximations. I show that advantageously chosen orderings of the observations can drastically improve the approximations. In addition to the permutation results\, automatic methods for grouping calculations of components of the approximation are introduced\, having the result of simultaneously improving the quality of the approximation and reducing its computational burden. In one common setting\, reordering combined with grouping reduces the Kullback-Leibler divergence from the target model by a factor of 80 and the computation time by a factor of 2 compared to ungrouped approximations with a default ordering. The claims are supported by theory and numerical results\, and details of implementation are provided\, including how to efficiently find the orderings and ordered nearest neighbors\, and how to use the approximations for prediction and conditional simulation. An application to uncertainty quantification in interpolations of space-time satellite data is presented. \n \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall \n
URL:https://stat-or.unc.edu/event/stor-colloquium-joe-guinness-north-carolina-state-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161017T153000
DTEND;TZID=America/New_York:20161017T163000
DTSTAMP:20210515T201830
CREATED:20160906T145045Z
LAST-MODIFIED:20170215T212349Z
UID:1884-1476718200-1476721800@stat-or.unc.edu
SUMMARY:STOR colloquium: Dirk Lorenz\, Braunschweig Technical University
DESCRIPTION:Dirk Lorenz \nBraunschweig Technical University \n \nProbabilistic Image Models and Extensions of the\nPerona-Malik Filter \nThe Perona-Malik model has been very successful at restoring images from noisy input. In this paper\, we show how the Perona-Malik model can be reinterpreted and extended using the language of Gaussian scale mixtures. Specifically\, we show how the expectation-maximization EM algorithm applied to Gaussian scale mixtures leads to the lagged-diffusivity algorithm for computing stationary points of the Perona-Malik diffusion equations. Moreover\, we show how mean field approximations to these Gaussian scale mixtures lead to a modification of the lagged-diffusivity algorithm that better captures the uncertainties in the restoration. Since this modification can be hard to compute in practice we propose relaxations to the mean field objective to make the algorithm computationally feasible. Our numerical experiments show that this modified lagged-diffusivity algorithm often performs better at restoring textured areas and fuzzy edges than the unmodified algorithm. As a second application of the Gaussian scale mixture framework\, we show how an efficient sampling procedure can be obtained for the probabilistic model\, making the computation of the conditional mean and other expectations algorithmically feasible. Again\, the resulting algorithm has a strong resemblance to the lagged-diffusivity algorithm. \n \n \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-dirk-lorenz-braunschweig-technical-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161031T153000
DTEND;TZID=America/New_York:20161031T163000
DTSTAMP:20210515T201830
CREATED:20160906T145152Z
LAST-MODIFIED:20170215T212349Z
UID:1886-1477927800-1477931400@stat-or.unc.edu
SUMMARY:STOR colloquium: Seyed Emadi\, University of North Carolina at Chapel Hill
DESCRIPTION:Impact of Delay Announcements in Call Centers: \nAn Empirical Approach \n \nWe undertake an empirical study of the impact of delay announcements on callers’ abandonment behavior and the performance of a call center with two priority classes. A Cox regression analysis reveals that in this call center\, callers’ abandonment behavior is affected by the announcement messages heard. To account for this\, we formulate a structural estimation model of callers’ (endogenous) abandonment decisions. In this model\, callers are forward-looking utility maximizers and make their abandonment decisions by solving an optimal stopping problem. Each caller receives a reward from service and incurs a linear cost of waiting. The reward and per-period waiting cost constitute the structural parameters that we estimate from the data of callers’ abandonment decisions as well as the announcement messages heard. The call center performance is modeled by a Markovian approximation. The main methodological contribution is the definition of an equilibrium in steady state as one where callers’ expectation of their waiting time\, which affects their (rational) abandonment behavior\, matches their actual waiting time in the call center\, and its characterization as the solution of a set of non-linear equations. A counterfactual analysis shows that callers react to longer delay announcements by abandoning earlier\, that less patient callers as characterized by their reward and cost parameters react more to delay announcements\, and that congestion in the call center at the time of the call affects caller reactions to delay announcements. \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall \n
URL:https://stat-or.unc.edu/event/stor-colloquium-seyed-emadi-university-of-north-carolina-at-chapel-hill/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161107T143000
DTEND;TZID=America/New_York:20161107T153000
DTSTAMP:20210515T201830
CREATED:20160906T145250Z
LAST-MODIFIED:20170215T212349Z
UID:1888-1478529000-1478532600@stat-or.unc.edu
SUMMARY:STOR colloquium: Ivo Adan\, Eindhoven University of Technology
DESCRIPTION:Ivo Adan \nTechnical University of Eindhoven and\nEURANDOM The Netherlands \n \nA rate balance principle \n \nWe introduce a rate balance principle for general (not necessarily Markovian) stochastic processes\, with special attention to birth-death like processes. This principle appears to be useful in deriving well-known\, as well as new results for various queueing systems. \n \n \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-ivo-adan-eindhoven-university-of-technology/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161114T143000
DTEND;TZID=America/New_York:20161114T153000
DTSTAMP:20210515T201830
CREATED:20160906T145348Z
LAST-MODIFIED:20170215T212348Z
UID:1890-1479133800-1479137400@stat-or.unc.edu
SUMMARY:STOR colloquium: CANCELLED
DESCRIPTION:This colloquium has been cancelled. We apologize for any inconvenience. \nSTOR colloquium: Vanja Dukic\, University of Colorado-Boulder
URL:https://stat-or.unc.edu/event/stor-colloquium-vanja-dukic-university-of-colorado-boulder/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161121T143000
DTEND;TZID=America/New_York:20161121T153000
DTSTAMP:20210515T201830
CREATED:20160906T145438Z
LAST-MODIFIED:20170215T212348Z
UID:1892-1479738600-1479742200@stat-or.unc.edu
SUMMARY:STOR colloquium: Stephen Becker\, University of Colorado-Boulder
DESCRIPTION:Stephen Becker \nUniversity of Colorado Boulder \n\nEfficient robust PCA algorithms for the GPU \n \nWe introduce the matrix completion problem and the similar robust PCA (RPCA) problem and discuss their relation to compressed sensing and some of their applications to collaborative filtering\, background detection in videos\, and neuroscience. We cover some standard algorithms to solve these problems\, including proximal gradient descent\, Frank-Wolfe/conditional-gradient\, and Burer-Monteiro splitting. A natural idea to speed up the algorithms is to run them on the GPU to take advantage of the many parallel threads of a GPU. This works well for some matrix completion algorithms\, but the RPCA algorithms do not parallelize well. This motivates our new algorithm for RPCA which is designed to run well on the GPU\, and we show results with major improvements over existing algorithms. \n \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-stephen-becker-university-of-colorado-boulder/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20161205T143000
DTEND;TZID=America/New_York:20161205T153000
DTSTAMP:20210515T201830
CREATED:20160906T145531Z
LAST-MODIFIED:20170215T212348Z
UID:1894-1480948200-1480951800@stat-or.unc.edu
SUMMARY:STOR colloquium: Elaine McVey\, TransLoc/Insightus
DESCRIPTION:Title: Commuting\, voting\, and the off-label use of data \n \nAbstract: The modern practice of data science often involves approaching datasets differently than in traditional statistics. Two analyses will be presented\, both of which use combinations of public datasets in expected and “off-label” ways. One addresses the need to simulate realistic commuting patterns\, and the other evaluates the effects of North Carolina voting site decisions. In the process of telling the stories of these two data science projects\, we’ll have an opportunity to reflect on the nature of data science. \n \nRefreshments will be served in the lounge at 3:30pm
URL:https://stat-or.unc.edu/event/stor-colloquium-elaine-mcvey-translocinsightus/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170130T143000
DTEND;TZID=America/New_York:20170130T153000
DTSTAMP:20210515T201830
CREATED:20170117T155930Z
LAST-MODIFIED:20170215T220855Z
UID:2574-1485786600-1485790200@stat-or.unc.edu
SUMMARY:STOR Colloquium: Santiago Balseiro\, Duke University
DESCRIPTION:Santiago Balseiro \nDuke University \nDynamic Mechanisms with Martingale Utilities \nWe study the dynamic mechanism design problem of a seller who repeatedly sells independent items to a buyer with private values. In this setting\, the seller could potentially extract the entire buyer surplus by running efficient auctions and charging an upfront participation fee at the beginning of the horizon. In some markets\, such as internet advertising\, participation fees are not practical since buyers expect to inspect items before purchasing them. This motivates us to study the design of dynamic mechanisms under successively more stringent requirements that capture the implicit business constraints of these markets. We first consider a “periodic individual rationality constraint\,” which limits the mechanism to charge at most the buyer’s value in each period. While this prevents large upfront participation fees\, the seller can still design mechanisms that spread a participation fee across the first few auctions. These mechanisms have the unappealing feature that they provide close-to-zero buyer utility in the first auctions in exchange for higher utility in future auctions. To address this problem\, we introduce a “martingale utility constraint\,” which imposes the requirement that from the perspective of the buyer\, the next item’s expected utility is equal to the present one’s. Our main result is providing a dynamic auction satisfying martingale utility and periodic individual rationality whose profit loss with respect to first-best (full extraction of buyer surplus) is optimal up to polylogarithmic factors. The proposed mechanism is a dynamic two-tier auction with a hard floor and a soft floor that allocates the item whenever the buyer’s bid is above the hard floor and charges the minimum of the bid and the soft floor. \nJoint work the Vahab Mirrokni (Google Research) and Renato Paes Leme (Google Research). \n \nThe paper is available at https://ssrn.com/abstract=2821261 \n\nSantiago Balseiro’s web page \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall \n
URL:https://stat-or.unc.edu/event/stor-colloquium-santiago-balseiro-duke-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170206T143000
DTEND;TZID=America/New_York:20170206T153000
DTSTAMP:20210515T201830
CREATED:20170117T160109Z
LAST-MODIFIED:20170215T220855Z
UID:2576-1486391400-1486395000@stat-or.unc.edu
SUMMARY:STOR Colloquium: Julie Ivy\, North Carolina State University
DESCRIPTION:Julie Ivy \nNorth Carolina State University \nTo be Healthy\, Wealthy\, and Wise: Using Decision Modeling to \nPersonalize Policy in Health\, Hunger Relief\, and Education \n \nDecision making to satisfy the basic human needs of health\, food\, and education is complex\, accomplishing this in such a way that solutions are personalized to the needs of the individual has been the focus of my research. This talk will present an overview of my research in the area of decision making under conditions of uncertainty with a focus on human-centric decision problems in three primary areas: health care\, humanitarian logistics\, and education. This research seeks to inform decision making with the goal of improving decision quality. To this end\, this research utilizes and develops theory in the areas of Markov decision processes (MDPs)\, semi-Markov decision processes (SMDPs)\, partially observable Markov decision processes (POMDPs)\, simulation\, and Bayesian decision analysis to address these real world problems. This research has made an impact on how researchers and practitioners address complex societal issues\, such as health disparities\, public health preparedness\, hunger relief\, student performance\, and personalized medical decision-making and has been funded by AHRQ\, CDC\, NSF\, Clinton Health Access Initiative\, and the UNC Cancer Center. \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-julie-ivy-north-carolina-state-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170213T143000
DTEND;TZID=America/New_York:20170213T153000
DTSTAMP:20210515T201830
CREATED:20170117T160202Z
LAST-MODIFIED:20170215T220854Z
UID:2578-1486996200-1486999800@stat-or.unc.edu
SUMMARY:STOR Colloquium: Peter Hoff\, Duke University
DESCRIPTION:Peter Hoff \nDuke University \nAdaptive FAB confidence intervals with constant coverage \n \nAbstract: Confidence intervals for the means of multiple normal populations are often based on a hierarchical normal model. While commonly used interval procedures based on such a model have the nominal coverage rate on average across a population of groups\, their actual coverage rate for a given group will be above or below the nominal rate\, depending on the value of the group mean. \n \nIn this talk I present confidence interval procedures that have constant frequentist coverage rates and that make use of information about across-group heterogeneity\, resulting in constant-coverage intervals that are narrower than standard t-intervals on average across groups. These intervals are obtained by inverting Bayes-optimal frequentist tests\, and so are “frequentist\, assisted by Bayes” (FAB). I present some asymptotic optimality results and some extensions to other multiparameter models\, such as linear regression. \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-peter-hoff-duke-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170220T143000
DTEND;TZID=America/New_York:20170220T153000
DTSTAMP:20210515T201830
CREATED:20170117T160252Z
LAST-MODIFIED:20170215T220854Z
UID:2580-1487601000-1487604600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Gudrun Johnsen\, University of Iceland
DESCRIPTION:Gudrun Johnsen \nUniversity of Iceland \n \nTitle: If you look close enough you can see a whole lot: Data collection and analysis of the Parliamentary Investigation Commission looking into the Icelandic Banking Collapse in 2008 \n \nAbstract: The combined collapse of Iceland’s three largest banks is the third largest bankruptcy in history and the largest banking system collapse suffered by any country in modern economic history\, relative to GDP. In 2008 the Icelandic Parliament installed a Special Investigation Commission (SIC) with exceptional data privileges to respond to the outcry of the Icelandic public and its demand to know why the banks failed. The SIC data collection\, data analysis and 2400-page report match few others in the public domain\, as Commissioners could cease all data considered necessary for the Investigation.
URL:https://stat-or.unc.edu/event/stor-colloquium-gundrun-johnson-university-of-iceland/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170301T143000
DTEND;TZID=America/New_York:20170301T153000
DTSTAMP:20210515T201830
CREATED:20170216T140137Z
LAST-MODIFIED:20170216T140137Z
UID:2767-1488378600-1488382200@stat-or.unc.edu
SUMMARY:STOR Colloquium: Mariana Olvera-Cravioto\, University of California-Berkeley
DESCRIPTION:Directed complex networks and ranking algorithms \nIn the first part of this talk I will discuss a family of inhomogeneous directed random graphs for modeling complex networks such as the web graph\, Twitter\, ResearchGate\, and other social networks. This class of graphs includes as a special case the classical Erdos-Renyi model\, and can be used to replicate almost any type of predetermined degree distributions\, in particular\, power-law degrees such as those observed in most real-world networks. I will mention during the talk the main properties of this family of random graphs and explain how its parameters can be used to represent important data attributes that influence the connectivity of nodes in the network. \nIn the second part of the talk I will explain how ranking algorithms such as Google’s PageRank can be used to identify highly influential nodes in a network\, and present some recent results describing the distribution of the ranks computed by such algorithms. This work extends prior work done for the directed configuration model to the new class of inhomogeneous directed random graphs mentioned above\, and provides a more natural way to model the relationship between highly ranked nodes and their attributes. If time allows\, I will mention some interesting stochastic simulation challenges related to this problem. \n \nRefreshments will be served in the lounge area of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-mariana-olvera-cravioto-university-of-california-berkeley/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170306T143000
DTEND;TZID=America/New_York:20170306T153000
DTSTAMP:20210515T201830
CREATED:20170117T160343Z
LAST-MODIFIED:20170221T172152Z
UID:2582-1488810600-1488814200@stat-or.unc.edu
SUMMARY:STOR Colloquium: Zhiyi Zhang\, UNC Charlotte
DESCRIPTION:Zhiyi Zhang \nUniversity of North Carolina at Charlotte \nTitle: Statistical Implications of Turing’s Formula \n \nAbstract: This talk is organized into three parts. \n \n\nTuring’s formula is introduced. Given an iid sample from a countable alphabet under a probability distribution\, Turing’s formula (introduced by Good (1953)\, hence also known as the Good-Turing formula) is a mind-bending non-parametric estimator of total probability associated with letters of the alphabet that are NOT represented in the sample. Many of its statistical properties were not clearly known for a stretch of nearly sixty years until recently. Some of the newly established results\, including various asymptotic normal laws\, are described.\n\n \n\nTuring’s perspective is described. Turing’s formula brought about a new perspective (or a new characterization) of probability distributions on general countable alphabets. The new perspective in turn provides a new way to do statistics on alphabets\, where the usual statistical concepts associated with random variables (on the real line) no longer exist\, for example\, moments\, tails\, coefficients of correlation\, characteristic functions don’t exist on alphabets (a major challenge of modern data sciences). The new perspective\, in the form of entropic basis\, is introduced.\n\n \n\nSeveral applications are presented\, including estimation of information entropy and diversity indices.
URL:https://stat-or.unc.edu/event/stor-colloquium-zhiyi-zhang-unc-charlotte/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170403T153000
DTEND;TZID=America/New_York:20170403T163000
DTSTAMP:20210515T201830
CREATED:20170117T160432Z
LAST-MODIFIED:20170328T132756Z
UID:2584-1491233400-1491237000@stat-or.unc.edu
SUMMARY:STOR Colloquium: Hongtu Zhu\, MD Anderson
DESCRIPTION:Hongtu Zhu \nUniversity of North Carolina at Chapel Hill\, and The University of Texas MD Anderson Cancer Center \n \nTitle: Statistical Challenges\, Opportunities\, and Strategies in Large-Scale Medical Studies \nWith the rapid growth of modern technology\, many biomedical studies have collected data across different sources (e.g.\, imaging\, genetics\, and clinical) in an unprecedented scale. The integration of such ultra high-dimensional data raises many statistical challenges\, rendering most existing statistical methods and old data platform no longer suitable and thus underscoring the great need for methodological developments from a rigorous perspective. To address these challenges\, I will highlight several key statistical opportunities and strategies in big data integration and analysis through several interrelated projects. \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:https://stat-or.unc.edu/event/stor-colloquium-hongtu-zhu-md-anderson/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170410T153000
DTEND;TZID=America/New_York:20170410T163000
DTSTAMP:20210515T201830
CREATED:20170117T160518Z
LAST-MODIFIED:20170410T131755Z
UID:2586-1491838200-1491841800@stat-or.unc.edu
SUMMARY:CANCELLED STOR Colloquium
DESCRIPTION:Philip Ernst \nRice University \n \nTitle: Yule’s “Nonsense Correlation” Solved! \n \nAbstract: In this talk\, I will discuss how I recently resolved a longstanding open statistical problem. The problem\, formulated by the British statistician Udny Yule in 1926\, is to mathematically prove Yule’s 1926 empirical finding of “nonsense correlation.” We solve the problem by analytically determining the second moment of the empirical correlation coefficient of two independent Wiener processes. Using tools from Fredholm integral equation theory\, we calculate the second moment of the empirical correlation to obtain a value for the standard deviation of the empirical correlation of nearly .5. The “nonsense” correlation\, which we call “volatile” correlation\, is volatile in the sense that its distribution is heavily dispersed and is frequently large in absolute value. It is induced because each Wiener process is “self-correlated” in time. This is because a Wiener process is an integral of pure noise and thus its values at different time points are correlated. In addition to providing an explicit formula for the second moment of the empirical correlation\, we offer implicit formulas for higher moments of the empirical correlation. The full paper is currently in press at The Annals of Statistics and can be found at http://www.imstat.org/aos/AOS1509.pdf.
URL:https://stat-or.unc.edu/event/stor-colloquium-philip-ernst-rice-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170417T153000
DTEND;TZID=America/New_York:20170417T163000
DTSTAMP:20210515T201830
CREATED:20170117T160611Z
LAST-MODIFIED:20170410T202935Z
UID:2588-1492443000-1492446600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Ilse Ipsen\, North Carolina State University
DESCRIPTION:Ilse Ipsen \nNorth Carolina State University\n \nRandomized Algorithms for Matrix Computations \n \nThe emergence of massive data sets\, over the past fifteen or so years\, has led to the development of Randomized Numerical Linear Algebra. Fast and accurate randomized matrix algorithms are being designed for applications like machine learning\, population genomics\, astronomy\, nuclear engineering\, and optimal experimental design. \nWe give a flavour of randomized algorithms for the solution of least squares/regression problems and\, if time permits\, for the computation of logdeterminants. Along the way we illustrate important concepts from numerical analysis (conditioning and pre-conditioning) and statistics (sampling and leverage scores).
URL:https://stat-or.unc.edu/event/stor-colloquium-ilse-ipsen-north-carolina-state-university/
LOCATION:Hanes Hall
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170503T153000
DTEND;TZID=America/New_York:20170503T163000
DTSTAMP:20210515T201830
CREATED:20170406T144008Z
LAST-MODIFIED:20170406T144008Z
UID:2891-1493825400-1493829000@stat-or.unc.edu
SUMMARY:STOR Colloquium: Jong-Shi Pang\, U. of Southern California
DESCRIPTION:Jong-Shi Pang \nUniversity of Southern California \nTitle: Structural Properties of Affine Sparsity Constraints \n \nAbstract: We introduce a new constraint system for sparse variable selection in statistical learning. Such a \nsystem arises when there are logical conditions on the sparsity of certain unknown model parameters that need \nto be incorporated into their selection process. Formally\, extending a cardinality constraint\, an affine sparsity \nconstraint (ASC) is defined by a linear inequality with two sets of variables: one set of continuous variables \nand the other set represented by their nonzero patterns. This paper studies an ASC system consisting \nof finitely many affine sparsity constraints. We investigate a number of fundamental structural properties \nof the solution set of such a non-standard system of inequalities\, including its closedness and the description \nof its closure\, continuous approximations and their set convergence\, and characterizations of its tangent \ncones for use in optimization. Based on the obtained structural properties of an ASC system\, we establish \nthe convergence of B(ouligand) stationary solutions when the ASC is approximated by surrogates of the \nell0 step function commonly employed in sparsity representation. Our study lays a solid mathematical \nfoundation for solving optimization problems involving these affine sparsity constraints through their \ncontinuous approximations. \n \nThis is joint work with Professor Hongbo Dong (Washington State University) and graduate student \nMiju Ahn (University of Southern California). \n \nRefreshments will be served in the lounge area of Hanes Hall at 3:00 pm
URL:https://stat-or.unc.edu/event/stor-colloquium-jong-shi-pang-u-of-southern-california/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170828T153000
DTEND;TZID=America/New_York:20170828T163000
DTSTAMP:20210515T201830
CREATED:20170811T203540Z
LAST-MODIFIED:20170822T182947Z
UID:3086-1503934200-1503937800@stat-or.unc.edu
SUMMARY:STOR Colloquium: Ion Necoara\, Polytechnic University of Bucharest
DESCRIPTION:Ion Necoara \nPolytechnic University of Bucharest \n \nConditions for linear convergence of (stochastic) first order methods \n \nFor convex optimization problems deterministic first order methods have linear convergence when the objective function is smooth and strongly convex. Moreover\, under the same conditions – smoothness and strong convexity – sublinear convergence rates have been derived for stochastic first order methods. However\, in many applications (machine learning\, statistics\, control\, signal processing) the strong convexity condition does not hold\, but the objective function still has a special structure. In this talk we replace the smoothness/strong convexity conditions with several other conditions\, that are less conservative\, for which we are able to prove that several (stochastic) first order methods are converging linearly. We also provide necessary conditions for linear convergence of (stochastic) gradient method. Finally\, we discuss several applications of these results (Lasso problem\, linear systems\, linear programming\, convex feasibility\, etc). \n \n \n \nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall \n
URL:https://stat-or.unc.edu/event/stor-colloquium-ion-necoara/
LOCATION:Hanes 120
CATEGORIES:STOR Colloquium
END:VEVENT
END:VCALENDAR