BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Statistics and Operations Research - ECPv4.5.2.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Department of Statistics and Operations Research
X-ORIGINAL-URL:http://stat-or.unc.edu
X-WR-CALDESC:Events for Department of Statistics and Operations Research
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20170922T153000
DTEND;TZID=UTC-4:20170922T163000
DTSTAMP:20170920T035501
CREATED:20170915T180711Z
LAST-MODIFIED:20170915T180711Z
UID:3189-1506094200-1506097800@stat-or.unc.edu
SUMMARY:Grad Student Seminar: Tianxiao Sun & Jonathan Williams
DESCRIPTION:Tianxiao Sun\nGlobally convergent Newton-type methods for convex optimization based\non smoothness structures\nWe study the smooth structure of convex functions by generalizing a powerful concept so-called self-concordance introduced by Nesterov and Nemirovskii to a broader class of convex functions. The proposed theory provides a mathematical tool to analyze both local and global convergence of Newton-type methods as long as the underlying functionals fall into our generalized self-concordant function class. First\, we introduce the class of generalized self-concordant functions\, which covers standard self-concordant functions as a special case. Next\, we establish several properties and key estimates of this function class. Then\, we apply this theory to develop Newton-type methods for solving a class of smooth convex optimization problems involving the generalized self-concordant functions. We provide an explicit step-size for the damped-step Newton-type scheme which can guarantee a global convergence without performing any globalization strategy. We also prove a local quadratic convergence of this method and its full-step variant without requiring the Lipschitz continuity of the objective Hessian. Then\, we extend our result to develop proximal Newton-type methods for composite convex minimization. We also achieve both local and global convergence without additional assumptions. Finally\, we verify our results via several numerical examples\, and compare them with existing methods.\nJonathan Williams\nA Bayesian Approach to Multi-state Modeling: Application to\nDementia Progression\nThis is joint work with Curtis Storlie\, and Terry Therneau\nthat is supported with funds from the Mayo Clinic.\nA multi-state model is a useful way of describing the development of a disease. Here\, a Hidden Markov Model (HMM) is specified to model dementia as it progresses from states associated with the buildup of amyloid plaque on the brain\, and the loss of cortical thickness. Both of these processes are known in the medical community to be strongly associated with dementia. In previous studies\, hard cutoff points were chosen to distinguish from states of high/low amyloid burden\, and high/low cortical thickness loss burden. However\, hard cutoff points for discretizing continuous measurements of biological processes are practically and philosophically problematic. Here\, an approach is proposed which is cutoff point agnostic. \nData on 4742 individuals from the Mayo Clinic Study of Aging (MCSA) are analyzed. A hierarchical Bayesian approach is taken to estimate the parameters of the HMM. A notable feature of the analysis is that time is treated as continuous\, and the infinitesimal generator matrix of the underlying Markov process is allowed to be time-inhomogeneous (as a function of an individual's age). A novel contribution is that in addition to the affect of age\, the affects of the covariates gender\, number of years of education\, and presence of an APOE4 allele on the infinitesimal transition rates are estimated. In this context\, these have never been estimated in the literature. \nIn addition to the new insights these estimates bring to the medical community\, a novel approach is illustrated for correcting a common bias in population-based studies. By 'population-based' study\, it is meant that not all (or even none) of the study participants are observed at a given baseline age. This often introduces a strong downward bias on the death rates because dead individuals typically are not recruited into a study. Finally\, standard R software is not capable of correcting for such biases\, and often doesn't include Bayesian estimation routines for a HMM.
URL:http://stat-or.unc.edu/event/grad-student-seminar-tianxiao-sun-jonathan-williams
CATEGORIES:Graduate Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171002T153000
DTEND;TZID=UTC-4:20171002T163000
DTSTAMP:20170920T035501
CREATED:20170811T203647Z
LAST-MODIFIED:20170905T165354Z
UID:3088-1506958200-1506961800@stat-or.unc.edu
SUMMARY:STOR Colloquium: Sercan Yildiz\, SAMSI
DESCRIPTION:Title: Polynomial Optimization with Sums-of-Squares Interpolants\nAbstract: Sums-of-squares certificates define a hierarchy of relaxations for polynomial optimization problems which are parameterized with the degree of the polynomials in the sums-of-squares representation. Each level of the hierarchy generates a lower bound on the true optimal value\, which can be computed in polynomial time via semidefinite programming\, and these lower bounds converge to the true optimal value under mild assumptions. However\, solving the semidefinite programs that arise from sums-of-squares relaxations poses practical challenges at higher levels of the hierarchy. First\, the sizes of these semidefinite programs depend quadratically on the number of monomials in the sums-of-squares representations. Second\, numerical problems are often encountered. In this talk\, we show that non-symmetric conic programming and polynomial interpolation techniques can be used to optimize efficiently over the sums-of-squares cone. Preliminary computational results indicate that our method compares favorably against standard approaches. The talk is based on joint work with David Papp.
URL:http://stat-or.unc.edu/event/stor-colloquium-sercan-yildiz
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171009T153000
DTEND;TZID=UTC-4:20171009T163000
DTSTAMP:20170920T035501
CREATED:20170811T203752Z
LAST-MODIFIED:20170811T203857Z
UID:3090-1507563000-1507566600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Patrick Wolfe\, Purdue
DESCRIPTION:
URL:http://stat-or.unc.edu/event/stor-colloquium-patrick-wolfe-purdue
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171016T153000
DTEND;TZID=UTC-4:20171016T163000
DTSTAMP:20170920T035501
CREATED:20170811T204013Z
LAST-MODIFIED:20170912T141117Z
UID:3093-1508167800-1508171400@stat-or.unc.edu
SUMMARY:STOR Colloquium: Srinagesh Gavirneni\, Cornell
DESCRIPTION:Title: Co-opetition in Service Clusters with Waiting-Area Entertainment\n\n Link to paper: Co-opetition-Service-Clusters\n\n\nAbstract: Unoccupied waiting feels longer than it actually is. Service providers operationalize this psychological principle by offering entertainment options in waiting areas. In a service cluster with a shared waiting space\, firms have an opportunity to cooperate in the investment for providing entertainment options while competing on other service dimensions. In this paper\, we develop a parsimonious model of co-opetition in a service cluster with shared entertainment options for waiting customers (e.g.\, a boardwalk). By comparing the case of co-opetition with two benchmarks (monopoly\, and duopoly competition)\, we demonstrate that a service provider\, which would otherwise be a local monopolist\, can achieve a higher pro fit by joining a service cluster and engaging in co-opetition: we numerically show that the average firm profit under co-opetition is 7.65% higher than under monopoly. Achieving such benefits\, however\, requires a cost-allocation scheme properly addressing a fairness-efficiency tradeoff. A pursuit of fairness may backfire and lead to even lower profits than under pure competition. We show that as much as co-opetition facilitates resource sharing in a service cluster\, it also heightens price competition. Furthermore\, as the intensity of price competition increases\, surprisingly\, service providers may opt to charge higher service fees\, albeit while providing a higher entertainment level.\n\n \n\nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:http://stat-or.unc.edu/event/stor-colloquium-srinagesh-gavirneni-cornell
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171023T153000
DTEND;TZID=UTC-4:20171023T163000
DTSTAMP:20170920T035501
CREATED:20170811T204130Z
LAST-MODIFIED:20170811T204130Z
UID:3095-1508772600-1508776200@stat-or.unc.edu
SUMMARY:STOR Colloquium: Jonathan Taylor\, Stanford
DESCRIPTION:
URL:http://stat-or.unc.edu/event/stor-colloquium-jonathan-taylor-stanford
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171106T153000
DTEND;TZID=UTC-4:20171106T163000
DTSTAMP:20170920T035501
CREATED:20170829T165906Z
LAST-MODIFIED:20170829T165906Z
UID:3128-1509982200-1509985800@stat-or.unc.edu
SUMMARY:STOR Colloquium; Xiao-Li Meng\, Harvard
DESCRIPTION:
URL:http://stat-or.unc.edu/event/stor-colloquium-xiao-li-meng-harvard
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171113T153000
DTEND;TZID=UTC-4:20171113T163000
DTSTAMP:20170920T035501
CREATED:20170829T170054Z
LAST-MODIFIED:20170829T170054Z
UID:3130-1510587000-1510590600@stat-or.unc.edu
SUMMARY:STOR Colloquium: Frank Permenter\, MIT
DESCRIPTION:
URL:http://stat-or.unc.edu/event/stor-colloquium-frank-permenter-mit
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171115T153000
DTEND;TZID=UTC-4:20171115T163000
DTSTAMP:20170920T035501
CREATED:20170822T182806Z
LAST-MODIFIED:20170831T192456Z
UID:3104-1510759800-1510763400@stat-or.unc.edu
SUMMARY:STOR Colloquium: Todd Kuffner\, Washington University in St. Louis
DESCRIPTION:Title: Philosophy of Science\, Principled Statistical Inference\, and Data Science\n\n \n\nAbstract: Statistical reasoning and statistical inference have strong historical connections with philosophy of science. In this talk\, the new paradigm of data-driven science is examined through comparison with principled statistical approaches. I will review the merits and shortcomings of principled statistical inference. The talk will feature a case study of post-selection inference\, recent progress regarding inference for black box algorithms\, and a survey of future challenges.\n\n \n\nRefreshments will be served at 3:00pm in the 3rd floor lounge of Hanes Hall
URL:http://stat-or.unc.edu/event/stor-colloquium-todd-kuffner-washington-university-in-st-louis
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171120T153000
DTEND;TZID=UTC-4:20171120T163000
DTSTAMP:20170920T035501
CREATED:20170829T170239Z
LAST-MODIFIED:20170829T170624Z
UID:3132-1511191800-1511195400@stat-or.unc.edu
SUMMARY:STOR Colloquium: Alex Belloni\, Duke
DESCRIPTION:
URL:http://stat-or.unc.edu/event/stor-colloquium-alex-belloni-duke
CATEGORIES:STOR Colloquium
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC-4:20171204T153000
DTEND;TZID=UTC-4:20171204T163000
DTSTAMP:20170920T035501
CREATED:20170829T170407Z
LAST-MODIFIED:20170829T170407Z
UID:3134-1512401400-1512405000@stat-or.unc.edu
SUMMARY:STOR Colloquium: Anru Zhang\, University of Wisconsin-Madison
DESCRIPTION:
URL:http://stat-or.unc.edu/event/stor-colloquium-anru-zhang-university-of-wisconsin-madison
CATEGORIES:STOR Colloquium
END:VEVENT
END:VCALENDAR