|
|
Fall 2008
PDE and Image Analysis Seminar
Speaker: Xingbin Pan (East China Normal University)
Time: Friday, August 29, 2008 at 4:00 pm
Place: MSB 117 (UConn - Storrs)
Abstract: This talk concerns a quasilinear system involving the curl operator. This system describes the Meissner state of anisotropic superconductors. The existence and the regularity of the weak solutions are discussed, and the asymptotic behavior of the solutions with small penetration depth is also examined. A comparison with the previous work jointly with Peter Bates for isotropic superconductors will be given
|
Actuarial Science Seminar
Speaker: Emiliano Valdez (University of Connecticut)
Time: Tuesday, September 2, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: In insurance and finance, one is occasionally faced with the need to examine the dependence structure of several random variables. A common approach is to express the relationship explicitly in functional form; statistical properties and inference are very well-developed in this area. A more general approach is to model the dependence through the specification of the joint multivariate distribution of the variables. Using the notion of copulas, we eliminate the effect of the marginal distributions which is believed to have nothing to do with the dependence structure of the random variables. In this talk, we introduce the concept of copulas and explore their fundamental properties. We illustrate some practical applications of copula modeling in insurance and finance, and we demonstrate how to calibrate copula models with data.
<Extra Information>
|
UConn Math Club
Speaker: Keith Conrad (UConn)
Time: Wednesday, September 3, 2008 at 5:30 pm
Place: MSB 319 (UConn - Storrs)
Abstract: In the mid-1600s, Fermat found a technique for studying solutions to equations in integers. His idea is that if an equation has a solution in integers, then you can try to find a solution that is smaller in some way. Understanding how to pass between smaller and larger solutions can either let you show there are no solutions or let you describe all the solutions in terms of some list of the smallest solutions.
The method of descent will be described through a couple of examples, having both algebraic and geometric meaning.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Algebra Seminar
Speaker: Ralf Schiffler (University of Connecticut)
Time: Tuesday, September 9, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Cluster Algebras are commutative algebras that have a special
combinatorial structure. Originally introduced to provide an algebraic
framework for the canonical basis in Lie theory, cluster algebras now have
connections to many different fields of mathematics, for example,
combinatorics of polyhedra, Poisson geometry, Teichmueller theory and
representations of finite dimensional algebras.
In the first talk, I will define cluster algebras (a not so easy task) and
explain some of the main results on cluster algebras. In the second talk,
I will explain how to use triangulated surfaces to do explicit
computations in cluster algebras.
|
Actuarial Science Seminar
Speaker: Jun Yan (Department of Statistics, University of Connecticut)
Time: Tuesday, September 9, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: Copulas have become a popular tool in multivariate modeling successfully applied in many fields. A good open-source implementation of copulas is much needed for more practitioners to enjoy the joy of copulas. This article presents the design, features, and some implementation details of the R package copula. The package provides a carefully designed and easily extensible platform for multivariate modeling with copulas in R. S4 classes for most frequently used elliptical copulas and Archimedean copulas are implemented, with methods for density/distribution evaluation, random number generation, and graphical display. Fitting copula-based models with maximum likelihood method is provided as template examples. With the classes and methods in the package, the package can be easily extended by user-defined copulas and margins to solve problems.
|
S.I.G.M.A. Seminar
Speaker: Keith Conrad (University of Connecticut)
Time: Wednesday, September 10, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: We will show how the degree of a rational function exhibits itself in several ways: analytic, algebraic, arithmetic, and maybe other ways that don't involve the letter a.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Tom Roby (UConn)
Time: Wednesday, September 10, 2008 at 5:30 pm
Place: MSB 403 (UConn - Storrs)
Abstract: One of the most powerful techniques for solving problems in
enumeration is to use a power series, called a generating
function, whose coefficients are the terms of a given sequence. A
generating function is essentially “a clothesline on which we hang the
terms of a sequence” (H. Wilf). Algebraic and analytic techniques can
then be applied to help us understand many properties of a
sequence--often to the point of deriving an explicit formula. We will
look at several examples, including famous sequences associated with
Fibonacci, Euler, and Catalan.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Logic Seminar
Algebra Seminar
Actuarial Science Seminar
Speaker: Ross Bowen (Phoenix Life Insurance Company)
Time: Tuesday, September 16, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: In this talk, I plan to discuss hedging at a high level with emphasis on the financial reporting implications and how this fits together with the entire company. I also plan to briefly discuss the actuarial profession, how the Society of Actuaries (SOA) and the American Academy of Actuaries (AAA) fit together, and what resources are available through them.
Comments: Free pizza will be served.
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: John Haga (University of Connecticut)
Time: Wednesday, September 17, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: The continued fraction expansion of e was first described by Euler, who proved that the expansion was accurate by solving a differential equation. This served as a proof that e is irrational. While developing a method to approximate power series by rational functions, Hermite's student Pade defined a class of integrals which, as we will see, can be used to offer an alternative proof as to the accuracy of the continued fraction expansion of e.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Steve Miller (UConn)
Time: Wednesday, September 17, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: The beauty of the Golden Ratio has inspired artists, architects and mathematicians for centuries. We will construct the Golden Ratio using straightedge and compass. We will also discuss continued fractions, pentagrams, the Fibonacci sequence and other representations of the ratio. The background for this talk is basic algebra and geometry.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Analysis and Probability Seminar
Speaker: Iddo Ben Ari (University of Connecticut)
Time: Friday, September 19, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: The homopolymer model I'll discuss consists of a family of Gibbs'
probability measures on the paths of continuous-time nearest-neighbor
random walk on the d-dimensional integer lattice. The measures are
indexed by two parameters real inverse temperature, eta, and non-
negative time, t. The measure corresponding to eta and t is
obtained from the symmetric random walk starting from the origin
through a Radon-Nykodim derivative, defined path-wise as the
exponent of eta times the occupation time at the origin up to time
t, divided by the normalizing partition function, which is the
expected value of these exponents. The object of study is the
asymptotic behavior of the measures as t tends to infinity. The
model exhibits a phase transition at some critical value of eta:
above the critical value is a "positive recurrent" phase, while
below it lies the "transient" phase. For dimensions three and
higher, there's also an intermediate phase corresponding to the
critical value of eta. The simplicity of the model allows to obtain
rather detailed description of each of the phases.
The results for positive eta were obtained in recent work by
Cranston and Molchanov, which motivated my work on negative eta
(non-trivial only in dimensions one and two).
|
Logic Seminar
Speaker: James Hunter
Time: Monday, September 22, 2008 at 4:45 pm
Place: Exley Science Center 618 (Wesleyan University)
Abstract: Traditional reverse mathematics deals with second-order objects: sets of natural numbers. This allows traditional reverse mathematics to use many techniques from computability theory and related fields. However, this also limits its ability to analyze topological statements, since all objects being discussed must be represented by second-order codes. The topological spaces examined in reverse mathematics have typically been of size continuum (so that points can be represented by sets of natural numbers) and countably-based (so that open sets can also be represented by sets of natural numbers).
Using a base theory defined in recent years by Kohlenbach, based on functionals rather than sets, one can start to perform higher-order mathematics. Extending Kohlenbach's base theory to allow topological spaces to consist of arbitrarily many "atoms," one can begin to rough out the reverse-mathematical strength of certain basic topological theorems.
|
Algebra Seminar
Speaker: Alvaro Lozano-Robledo (University of Connecticut)
Time: Tuesday, September 23, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: An elliptic curve E over Q is a smooth projective curve of genus 1 defined over the rationals (and with at least one rational point, say O). Using an old geometric method, the set of rational points on E can be given an abelian group structure, with O acting as the neutral element. The Mordell-Weil theorem shows that (E(Q),+) is a finitely generated abelian group. Therefore, E(Q) is isomorphic to TxZ^r, where T is a finite group (the torsion subgroup), Z are the integers and r is a non-negative integer, called the rank of E/Q.
In the first of these two talks I will provide a summary of what is known about the torsion subgroup of E/Q, and also about elliptic curves over other number fields. In particular, I will describe some joint work with Ben Lundell, where we provide bounds for the growth of the torsion subgroup as the base field of a fixed curve is enlarged (i.e. if E/Q is an elliptic curve defined over Q, and K is a number field, what is the torsion subgroup of E(K)?).
In the second talk I will concentrate on questions about the rank of elliptic curves and, once again, I will discuss how the rank may change when the base field is enlarged to larger number fields.
|
Actuarial Science Seminar
Speaker: Richard Derrig (OPAL Consulting)
Time: Tuesday, September 23, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: Rate regulation has a long history in insurance markets. In many states an important goal of regulation is to reduce price variation across purchasers, and specifically to reduce price levels for high-risk purchasers. That feature of rate regulation leads to price cross-subsidies from low-risk purchasers to high-risk purchasers. Consumers who are charged higher prices in order to finance cross-subsidies to high-risks may be less likely to purchase insurance and to reduce participation in insured activities. These adverse selection effects will lead to a higher proportion of high-risk consumers and a higher proportion of insurance purchased by high-risks. In addition, because cross-subsidies reduce the links between insurance risk and insurance prices, all consumers face reduced incentives for loss prevention and safety investments due to moral hazard.
The article tests the hypothesis that insurance price subsidies lead to higher insurance cost growth. To squarely focus on the impact of regulatory price subsidies rather than that of price regulation more generally, the paper makes use of data from the Massachusetts private passenger automobile insurance market. Cross-subsidies were explicitly built into the rate structure through rules that limit rate differentials and differences in rate increases across driver rating categories. Two approaches were taken to study the potential loss cost reaction to the inefficient Massachusetts cross-subsidies that began in systematic form in 1977 and continue through 2007. The first approach compared Massachusetts to all other states on demographic, regulatory and liability coverage levels. Loss cost levels that were 44 to 50 percent above the expected level were found for Massachusetts during the 1978-1995 periods when premiums charged were those fixed by the state and included explicit cross subsidies from low risk drivers to high risk drivers. A second approach considered changing cost levels across Massachusetts by studying loss cost changes by town and relating those changes to subsidy providers and subsidy receivers. Subsidy data for 1999-2007, with underlying accident year data for 1993-2004, showed a significant and positive (relative) growth in loss costs for towns that were subsidy receivers in line with the theory of underlying incentives for adverse selection and moral hazard.
Comments: Click below for slides and report
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: Asher Kach (University of Connecticut)
Time: Wednesday, September 24, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Most currency coins are not fair in the sense that heads and
tails will not arise with equal probability. The first half of the talk
will be a discussion of how efficiently (if at all) one can simulate a
"fair" coin from a "biased" coin and the algorithms one might use to do
so.
The second half of the talk will be a discussion of the resulting
sequences. For example, intuitively the sequence should be "random" if a
fair coin is used, but "less random" if a biased coin is used. Various
notions of mathematical randomness will be introduced and partially
formalized in not only betting [martingale] contexts, but also within
analytic contexts (effective measure theory) and descriptive contexts
(Kolmogorov complexity).
Of course, no specific mathematical background is presumed.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Oscar Levin (UConn)
Time: Wednesday, September 24, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: The Four Color Theorem states that every map can be colored with at most four colors in such a way that no bordering countries share the same color. The theorem was first proved in 1976. However, in 1879 Alfred Kempe gave a ''proof'' whose mistake wasn't noticed for 11 years! In this talk we will build up the necessary graph theory to convince ourselves that Kempe's proof is correct, and then see why it is not.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Analysis and Probability Seminar
Speaker: Mike Keane (Wesleyan University)
Time: Friday, September 26, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: In this lecture we shall treat the recurrence (or possible transience;
there are open questions here) of once reinforced random walks on the
integers and on products of integers with finite segments of integers,
called ladders. Concerning once reinforcement on the integers, where
we can prove that no matter what the strength of the reinforcement
(or weakening) is, such random walks are recurrent. Then we shall
discuss once reinforcement on ladders. If there are only two copies
of the integers, then Sellke has
proved that once reinforced random walk is recurrent for any positive
reinforcement, and together with Feiden we have now a proof that this
remains true for negative reinforcement (i.e. weakening). Both questions
are still open for ladders of widths greater than two, although there
are positive results for some values of positive reinforcement due to
Sellke (low values of positive reinforcement) and Vervoort (high values
of positive reinforcement). Of course, it is expected that for any width
and any reinforcement, positive or negative, random walk is recurrent,
and even if we consider the case of two dimensions, i.e. infinite
ladders we expect recurrence. However, the latter problem seems to be
well beyond reach using current techniques. The last part of the lecture
deals with a new conjecture of mine, concerning the
recurrence-transience dichotomy for once reinforced random walks on
countably infinite, locally finite, connected graphs. All of the
concepts are easily accessible for a general mathematical audience.
Comments: Should be accessible to a general math audience
|
Logic Seminar
Speaker: Russell Miller (Queens College, CUNY)
Time: Monday, September 29, 2008 at 4:45 pm
Place: Exley Science Center 618 (Wesleyan University)
Abstract: Computable model theory has always restricted itself to the study of countable structures. The natural numbers form the standard domain and range for partial computable functions, due to the finiteness of the programs and computations in the Turing model. In this talk we continue to use this standard model of computation, but in such a way as to allow consideration of uncountable structures. We no longer aspire to describe a structure S globally. Instead, we content ourselves with a local description: a list of the finitely-generated substructures of S, up to isomorphism. Initially we require that these substructures should be computable, and that the list of them be given effectively. If this can be done, then S is said to be locally computable. (Such an S must have only countably many finitely-generated substructures, up to isomorphism.) Then we consider more sophisticated effective descriptions of S, by naming embeddings among the substructures which reflect the inclusions in S. We will prove several theorems about locally computable structures, but since the notion of local computability is extremely new, much of this talk will be devoted to definitions and examples.
|
Algebra Seminar
Speaker: Alvaro Lozano-Robledo (University of Connecticut)
Time: Tuesday, September 30, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: An elliptic curve E over Q is a smooth projective curve of genus 1 defined over the rationals (and with at least one rational point, say O). Using an old geometric method, the set of rational points on E can be given an abelian group structure, with O acting as the neutral element. The Mordell-Weil theorem shows that (E(Q),+) is a finitely generated abelian group. Therefore, E(Q) is isomorphic to TxZ^r, where T is a finite group (the torsion subgroup), Z are the integers and r is a non-negative integer, called the rank of E/Q.
In the first of these two talks I will provide a summary of what is known about the torsion subgroup of E/Q, and also about elliptic curves over other number fields. In particular, I will describe some joint work with Ben Lundell, where we provide bounds for the growth of the torsion subgroup as the base field of a fixed curve is enlarged (i.e. if E/Q is an elliptic curve defined over Q, and K is a number field, what is the torsion subgroup of E(K)?).
In the second talk I will concentrate on questions about the rank of elliptic curves and, once again, I will discuss how the rank may change when the base field is enlarged to larger number fields.
|
Actuarial Science Seminar
Speaker: Gail Hall (ACTEX Publications)
Time: Tuesday, September 30, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: The Society of Actuaries (SOA) Education System has undergone considerable change over the last three years. Reorganization of content, new topics, and new delivery systems have all been a part of the change. The latest is the introduction of a continuing professional development requirement for all members beginning in 2009. In this session, we will summarize all of the changes that have occurred since 2005 and then explore the new professional development requirement.
<Extra Information>
|
UConn Math Club
Speaker: Paul Ellis (UConn)
Time: Wednesday, October 1, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: I will discuss three invariants for knots, one combinatorial and two
algebraic. We will be able to use these invariants to distinguish between
some common knots. Only a little familiarity with polynomials will be
assumed. I will define everything else, including what I mean by a knot.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Analysis and Probability Seminar
Speaker: Mihai Pascu (University of Brasov)
Time: Friday, October 3, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: A well-known conjecture of R. Laugesen and C. Morpurgo asserts
that the diagonal element of the Neumann heat kernel of the unit disk is a radially increasing function. In this talk, we use probabilistic arguments to settle this conjecture. As an application, we obtain a new proof of the Hot Spots conjecture of J. Rauch in the case of the unit disk and we present some extensions.
Comments: Speaker is a former UConn graduate student.
|
Logic Seminar
Speaker: Philip Scowcroft (Wesleyan University)
Time: Monday, October 6, 2008 at 4:45 pm
Place: Exley Science Center 618 (Wesleyan University)
Abstract: In 1976 Elliott showed that approximately finite-dimensional C*-algebras (AF algebras) could be classified up to isomorphism by certain countable partially ordered Abelian groups (dimension groups). It turns out that one of these groups is algebraically (existentially) closed in another just in case the corresponding algebras, viewed as metric structures, enjoy the same relation.
|
Actuarial Science Seminar
Speaker: James Bridgeman (University of Connecticut)
Time: Tuesday, October 7, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: A regime-switching stochastic model with randomized regime parameters creates a more plausible set of extreme paths for stress-testing than do the usual risk-neutral interest rate models. Generalizing the Black-Karasinski model by randomizing the mean reversion target provides an example. Calibrating to historical data using previously derived relationships, Monte Carlo calculations illustrate the characteristics of the model.
|
S.I.G.M.A. Seminar
Speaker: Avraham Bourla (University of Connecticut)
Time: Wednesday, October 8, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Most of us know the statement of the four color theorem. It answers the question of what is the maximal number of colors needed to paint any planar map such that neighboring regions are painted with different colors. In this talk we'll generalize this question from the plane to the realm of all surfaces. We will define the chromatic number topological invariant and relate it to other invariants such as the Euler charateristic and the homology groups. We'll have fun coloring maps on the torus, mobius strip, projective plane etc. (you can leave your color box at home).
Comments: Free Donuts (courtesy of GSS)
Additional Comments: Note the normal start time.
|
UConn Math Club
Speaker: Richard Bass (UConn)
Time: Wednesday, October 8, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: One of the simplest models of gambling is coin tossing: toss a fair
coin over and over. If it comes up heads, you win $1; if tails, you
lose $1. First we'll talk about boring stuff like how long it will take to win
a billion dollars by gambling, and then
go on to more interesting things like the law of averages, the law
of the iterated logarithm, the arc sine law, the most visited site,
and things like that. For example, in the picture if n is the number of
the toss and Sn is your fortune at time n,
the dotted line is always asymptotic to the function
f(n) = √2n log log n.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Colloquium
Speaker: Wesley Calvert (Murray State University)
Faculty Sponsor: Reed Solomon
Time: Thursday, October 9, 2008 at 4:00 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Early in the development of twentieth-century mathematics, van de Waerden and others asked questions about the existence of "explicit" solutions to many problems whose standard proofs gave none. With the introduction of precise definitions of algorithms in the 1930's, these questions were investigated for some time before the revival in recent decades of "computational mathematics."
In the present talk I will attempt to describe relationships between the classical logical discipline of "computable mathematics" and the modern field of "computational mathematics." Several non-equivalent definitions of computation will be used. Examples will include fields, rings of integers, homotopy groups, the classification of manifolds, and Serre's conjecture on free modules.
|
Analysis and Probability Seminar
Speaker: Naotaka Kajino (Kyoto University)
Time: Friday, October 10, 2008 at 2:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: On a self-similar set like the Sierpinski gasket or the Sierpinski carpet,
a canonical self-similar diffusion process (the `Brownian motion') is known
to exist and we may regard its infinitesimal generator as the `Laplacian'
on that space.
We are interested in asymptotic behaviors of the eigenvalues of this Laplacian.
The case of the Sierpinski gasket is relatively easy since it is finitely
ramified, that is, each pair of adjacent cells intersect on a finite set.
But since the Sierpinski carpet is infinitely ramified, there had been almost
no result until a recent result by B. M. Hambly (preprint 2006).
In the talk we start with a brief description of the Brownian motion, the
Laplacian and spectral analysis on the Sierpinski gasket. After that
we will discuss how we can make use of the diffusion and the heat kernel
estimate to `avoid' (not `overcome') the difficult arising for the case of
the Sierpinski carpet.
Comments: Special bonus seminar - note not usual time!
|
Analysis and Probability Seminar
Speaker: Conrad Plaut (University of Tennessee)
Time: Friday, October 10, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: I'll discuss joint work with Valera Berestovskii (Omsk Branch of the Sobolev Institute and University of Notre Dame) concerning the Uniform Universal cover (UU-cover) and the Covering R-tree. The UU-cover exists and has nice functorial, lifting, and universal properties for a large category of uniform spaces, including all connected locally compact groups, all geodesic spaces (which of course includes many fractals and the Hawiian earring), and even pathological favorites like the Topologist's Sine Curve. In many cases the UU-cover is not well-suited to geometry or analysis, but I'll give a nice application to rectifiable curves that leads to the construction of the covering R-tree. The covering R-tree exists for every geodesic space, is unique, and is in some sense the "mother of all metric covers".
Comments: Combined seminar with Geometry-Topology!
|
Algebra Seminar
Speaker: Martin Nikolov (University of Connecticut)
Time: Tuesday, October 14, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: We develop a relative trace formula for the pair
(GSp(4),SO(4,E/F)) of the symplectic group of similitudes GSp(4) and
special orthogonal group SO(4,E/F) associated
to a quadratic extension E of the base field F. Using this trace formula
we describe the automorphic representations of PGSp(4) with a non-zero
period for SO(4) and a nonzero Fourier coefficient for a generic character
of the unipotent radical of the Siegel parabolic subgroup, as the image
of a functorial lifting of automorphic representations from PGL(2) to
PGSp(2).
|
Actuarial Science Seminar
Speaker: Tom Campbell (Hartford Life)
Time: Tuesday, October 14, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: The National Association of Insurance Commissioners (NAIC) is in the midst of a project to modernize life insurance statutory reserves and risk-based capital in the Untied States. The new method, referred to as principle-based approach, is based on a set of core principles and uses risk analysis and risk management techniques to capture underlying life insurance and annuity risks more accurately than the current rule-based, one-size-fits-all approach. The American Academy of Actuaries has taken the lead in developing a new framework for this modernization. This presentation will explore these efforts and discuss the framework itself, the players involved in this initiative, how the approach is used, and the status of particular components of this approach.
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: Catherine Ross (University of Connecticut)
Time: Wednesday, October 15, 2008 at 5:00 pm
Place: MSB 118 (UConn - Storrs)
Abstract: We are very happy to welcome Dr. Catherin Ross, Associate Director of ITL and Director of the UConn TA Program, who has graciously offered to run a two part workshop on writing teaching statements. At the first meeting, we will go over some basic principles for writing effective teaching statements. The second meeting will be for anyone who wishes to get feedback on a teaching statement they have written.
A good teaching statement is essential in the application process of almost all academic jobs. For more information on the importance of teaching statements, check out this article from the AMS Notices: http://www.math.uconn.edu/TAProgram/CareerInformation/Teaching-Statements-DEREK_BRUFF.pdf
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Warren Johnson (Connecticut College)
Time: Wednesday, October 15, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: Few if any mathematicians working today would list calculus
as one of their research areas. Are there any interesting calculus
theorems that are still to be discovered? Several years ago I found
some peculiar generalizations of Leibniz's rule for the nth derivative
of a product. I will discuss Leibniz's rule and these extensions, and
tell the story of what happened when I tried to find out if they were
new or not.
Comments: Free Refreshments USG Funded
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Analysis and Probability Seminar
Speaker: Dapeng Zhan (Yale)
Time: Friday, October 17, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: Abstract: I will talk about the proof of the reversibility conjecture and
duality conjecture of the stochastic Loewner evolution (SLE) introduced
by O. Schramm. The reversibility conjecture is proposed by Rohde and
Schramm, which says that for $kappa in [0,8]$, the reversal of the
chordal SLE$_kappa$ curve in $D$ from $a$ to $b$ agrees with the chordal
SLE$_kappa$ curve in $D$ from $b$ to $a$ in distribution. The duality
conjecture by Duplantier claims that for $kappa>4$, the outer boundary of
an SLE$_kappa$ hull locally looks like an SLE$_{16/kappa}$ curve. My
results are that the reversibility conjecture holds for $kappain(0,4]$,
and the duality conjecture holds for all $kappa>4$. And these two
conjectures are proved using the same technique.
|
Logic Seminar
Speaker: Zoe Chatzidakis (University Paris VII)
Time: Monday, October 20, 2008 at 5:15 pm
Place: Leet Oliver (Yale Mathematics Department 214 (Yale University)
Abstract: (Joint with E. Hrushovski) Let K be a field, V a variety, and f:V to V a morphism, both defined over K. Assume that there is a subfield k of K, another variety W with a morphism g:W to W, and a rational map h:W to V such that h(W) is Zariski dense in V, and fh=hg. One can then think of (V,f) as a quotient of (W,g) by the rational map h.
I will discuss how this setting translates in terms of difference fields, and what certain tools from model theory can tell us. One of the results one obtains is that (V,f) in turn has a non-trivial quotient (V',f') which is defined over the small field k.
Comments: Note unusual time and location!
|
Algebra Seminar
Speaker: Martin Nikolov (University of Connecticut)
Time: Tuesday, October 21, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: We develop a relative trace formula for the pair
(GSp(4),SO(4,E/F)) of the symplectic group of similitudes GSp(4) and
special orthogonal group SO(4,E/F) associated
to a quadratic extension E of the base field F. Using this trace formula
we describe the automorphic representations of PGSp(4) with a non-zero
period for SO(4) and a nonzero Fourier coefficient for a generic character
of the unipotent radical of the Siegel parabolic subgroup, as the image
of a functorial lifting of automorphic representations from PGL(2) to
PGSp(2).
|
Actuarial Science Seminar
Speaker: Craig Simms (Vantis Life Insurance Company)
Time: Tuesday, October 21, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: The CMO of a small, niche life insurer will present the variables that must be considered when designing and pricing life insurance products. Special attention and discussion will be paid to creating product features and riders that appeal to the target market, how the product is quoted and submitted and how pricing may or may not be an important variable in effectively competing in the marketplace.
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: Matthew Jura (University of Connecticut)
Time: Wednesday, October 22, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract:
Most of us are familiar with the chromatic number of a graph. We consider a different way to color a graph G = (V, E), known as the coloring number of G -- written Col(G). We say Col(G) is the least cardinal κ for which there is a well ordering of the vertex set V in which every vertex is joined to fewer than κ smaller vertices. We will start with some examples, and then we will use diagonalization (a standard technique of computability theory) to show that there is a computable forest G = (V, E) such that no computable well ordering of V realizes any finite coloring number for G. No knowledge of computability theory or graph theory is assumed.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Alvaro Lozano-Robledo (UConn)
Time: Wednesday, October 22, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: In this talk we will concentrate on the life of a great mathematician,
Leonardo ''Bigollo'', and the life of a great problem, the Congruent Number Problem. An integer n is called a “congruent number” if some right triangle with rational sides has area n. For example,
the 3-4-5 right triangle has area 6, so 6 is a congruent number.
Less obviously, the right triangle with sides 3/2, 20/3, 41/6
has area 5, so 5 is a congruent number. Fermat proved 1 and 2 are not congruent numbers.
Deciding when an integer n is a congruent number is a hard problem, and we will discuss some of what is known and unknown about it.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Actuarial Science Seminar
Speaker: Hubert Mueller (Towers Perrin)
Time: Tuesday, October 28, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: This session will discuss the latest trends in European Embedded Value (EEV) and Market-Consistent Embedded Value (MCEV) reporting, both globally and in the US. Topics covered will include: (1) How EEV and MCEV differ from traditional Embedded Value; (2) Latest CFO Forum Recommendations; (3) Link to Economic Capital / performance measurement; and (4) Use in making business decisions.
Comments: Click below for slides and article
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: Russell Prime (University of Connecticut)
Time: Wednesday, October 29, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: The title might not make any sense (I'm positive it is at best
wildly ungrammatical), but the point is that number theory can be done in
function fields (as opposed to classical number fields). What does that
mean? The familiar integers share several key features with F[T], the
ring of polynomials with coefficients in a finite field F. Our goal will
be to see what these features are, and how they lead to analogues of some
famous classical theorems in this new setting. We'll talk about things
ranging from Euler's phi function, to the Prime Number Theorem, to the
fate of the Riemann Hypothesis for function fields (spoiler: it's true!).
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Caleb Shor (Western New England College)
Time: Wednesday, October 29, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: To read information from a compact disc, a CD player shines a light on a
revolving CD and measures the intensity of the reflection. Variations
in intensity (high and low) correspond to sequences of 0s and 1s, which
form the data on the disc. If a disc has a scratch, then the CD player
cannot measure the light intensity correctly, but often the sound quality is not affected. How is that possible?
The answer is with error-correcting codes. In this talk, we will see
what error-correcting codes are, discuss their properties, and
understand how some commonly used real-world codes are created.
Comments: Free Refreshments USG Funded
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Colloquium
Speaker: Charles F. Van Loan (Joseph C. Ford Professor of Engineering, Cornell University )
Faculty Sponsor: Michael Neumann
Time: Thursday, October 30, 2008 at 4:00 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Historically, matrix-based computational thinking has spread from specific application areas to the larger scientific community. Consider how the numerical PDE community powered the development of the first sparse matrix solvers, or how the control and signal processing engineers instigated the development of structured/parallel matrix computations. Likewise, I believe that the emerging field of tensor computations will advance through the development of numerical methods from application areas such as quantum chemistry, where the dimension of the data objects make it the perfect venue for inventing general multilinear methods that scale. Among the topics to be discussed are (a) the challenges of multilinear optimization, (b) "low rank" tensor approximation, (c) the idea of tensor networks, and (d) the need for block tensor data structures.
Comments: This a joint colloquium with CSE
|
Analysis and Probability Seminar
Speaker: Marius Ionescu (University of Connecticut)
Time: Friday, October 31, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: Let $X$ be a compact space. In this talk a Markov operator is a unital
positive map $P: C(X) o C(X)$. For example homeomorphisms and local
homeomorphisms on $X$ define Markov operators. In will present a class
of $C^*$-algebras associated with such Markov operators. I will discuss
in particular $C^*$-algebra arising from Markov operators on fractals
and other self-similar sets. I will also make the case that these
algebras cover a large class of $C^*$-algebras. This talk is based on
joint work with Paul S. Muhly.
|
Logic Seminar
Speaker: Jared Corduan (Dartmouth)
Time: Monday, November 3, 2008 at 4:45 pm
Place: Exley Science Center 618 (Wesleyan University)
Abstract: A partially ordered set P has the (n,k)-Ramsey property if every coloring of the n-chains of P with k colors has a monochromatic suborder isomorphic to P. In their paper Reverse mathematics, computability, and partitions of trees, Chubb, Hirst, and McNicholl analyzed the proof theoretic strength of Ramsey's theorem for the complete binary branching tree. I will talk about joint work with Marcia Groszek and Joseph Mileti on Ramsey's theorem for arbitrary trees and answers to some questions raised by Chubb, Hirst, and McNicholl.
|
Algebra Seminar
Speaker: Jon Axtell (University of Connecticut)
Time: Tuesday, November 4, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: In his paper, "Vertex operator algebras associated to type B affine Lie algebras on admissible half-integer levels", O. Perse gives a classification of modules for a class of VOAs corresponding to type B affine Lie algebras. In the talk, we will first give the definition of a vertex operator algebra, and look at examples associated with affine Lie algebras. We will then discuss the classification mentioned above for the type B case, and possibly show how to extend this classification to the type G case.
|
Actuarial Science Seminar
Speaker: Edward (Jed) Frees (University of Wisconsin - Madison)
Time: Tuesday, November 4, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: The insurance industry is known to have high operating expenses in the financial services sector. Insurers, investors and regulators are interested in models to understand the behavior of expenses. Although many have considered expenses in the insurance and risk management literature, current practice is to ignore skewness, occasional negative values as well as their intertemporal dependence.
Addressing these three features, this paper develops a longitudinal model of insurance company expenses that can be used to predict, to identify unusual behavior, and to measure firm efficiency. Specifically, we use a three parameter asymmetric Laplace density to model the marginal distribution of insurer's expenses in each year. As a function of explanatory variables, the location parameter allows us to analyze insurer's expense in the light of firm's characteristics and provides a natural link to quantile regression. Copula functions are used to model the intertemporal dependence of expenses over time. In this way, we introduce a general approach to constructing quantile regression models for longitudinal data.
The analysis is performed using property-casualty insurance company data from the National Association of Insurance Commissioners for years 2001-2006. Due to the long-tailed nature of insurer's expenses, two alternative methods are proposed to improve the performance of the longitudinal quantile regression model: rescaling and transformation. Predictive densities are derived that allow one to compare the predictions for individual insurers in a hold-out-sample. Both predictive models are shown to be reasonable with the rescaling method outperforming the transformation method. Compared with standard longitudinal models, our model is shown to be superior in identifying insurers' unusual behavior.
Comments: Click below for slides and article
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: Catherine Ross (University of Connecticut)
Time: Wednesday, November 5, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: This second meeting of the two part workshop on writing effective teaching statements will be for anyone who wishes to get feedback on a teaching statement they have written. Everyone attending should bring a few copies of their teaching statement draft. We will work in groups to critique the drafts.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Oleg Mikitchenko (Connecticut College)
Time: Wednesday, November 5, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: Let f(x,y)= ax+by+cxy+.. be a polynomial in two variables without a constant term. If one of the linear coefficients a or b is not zero, then the equation f(x,y)=0 has only one solution y=y(x) (or x=x(y)) passing through the origin, and this solution can be written as a standard power series. For instance, if 5x + y - y2 + 3x2y = 0 then
near the origin y = -5x + 25x2 - 235x3 + ....
However, if a and b are both zero, then the equation f(x,y)=0 will generally have several solution curves passing through the origin, and these curves can be represented as series in fractional powers of x.
For instance, if y2 + 2x2y - x3 = 0 then y = ±(x3/2 - x2 + ½x5/2 + ...).
A polygon in the plane, called the Newton polygon, can be used to obtain these fractional power series solutions. In this talk we will show how the Newton polygon for the equation f(x,y)=0 is constructed and used.
Comments: Free Refreshments USG Funded
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Mathematics Education
Speaker:
Time: Thursday, November 6, 2008 at 3:30 pm
Place: MSB 120 (UConn - Storrs)
Abstract: A conversation on the role of Mathematics Education in our
department. Anyone interested in mathematics
education is welcome.f
Additional Comments: Wine and cheese will be served
|
Algebra Seminar
Speaker: Russell Prime (University of Connecticut)
Time: Tuesday, November 11, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Gauss posed a problem about the average value of class numbers
of quadratic number fields, and by Dirichlet's class number formula, this
can be generalized to a question about the average value of L(s,chi) over
quadratic characters chi. In part I of the talk, we will discuss the
results that address this problem, and introduce the analogous ideas in
the function field setting. In part II of the talk, I will talk about my
current work in averaging L-functions over function fields, and discuss a
proof of a result concerning the average of L(s,chi) for quadratic chi
with irreducible modulus.
|
Actuarial Science Seminar
Speaker: Hui Shan (University of Connecticut)
Time: Tuesday, November 11, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: The secondary market of life insurance has demonstrated spectacular growth in past years. According to a study published by Conning Research & Consulting in October 2007, the total life insurance face amount transferred in 2006 was estimated to be about $6.1 billion, which reflects more than 30 percent compound growth from 2002. For investors who engage in the secondary market of life insurance, it becomes more and more critical to understand the risk, price the risk, track the performance and adjust assumptions to account for experience. A rigorous actuarial pricing and performance tracking process helps gain a competitive edge in this arena. This presentation introduces a few techniques to help major players such as investors, insurers and reinsurers address the inefficiencies and challenges they face, and provide solutions.
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: James Bridgeman (University of Connecticut)
Time: Wednesday, November 12, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: The application of quantitative models and techniques to finance and investing has been credited with helping to transform the world we live in (circumspice … think of cellular, cable, wireless, consumer electronics, web commerce and entertainment, the lifting of many hundreds of millions of human beings out of poverty in just 30 years, the ability to turn home ownership into cash … woops!). It is currently being blamed for financial catastrophe. What mathematics could possibly be involved in such a range of effects? We’ll sketch some of the foundations, why it takes a fair hand at mathematics to deal with them, and try to indicate how they get implicated in so many real world effects.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Jennifer Beineke (Western New England College)
Time: Wednesday, November 12, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: Graph theory can provide an entertaining analysis of certain games and
puzzles. Using elementary results, we will explore brainteasers such as
Instant Insanity, Dots and Boxes, and Paradoxical Pennies. That
should be preparation enough to set us off on a mathematical sort of
safari.
Here is an example of one problem we'll discuss which you might think about before the talk:
In one room, you have 100 prisoners, and in another room, you have 100 jars lined up on a table. Each prisoner's name is in exactly one of the 100 jars. One by one, the prisoners are led into the room with the jars. Each prisoner may look in at most 50 jars, but must leave the room exactly as he found it, and is not allowed to communicate with others.
The prisoners are allowed to strategize in advance, and they are going to need it, because unless every single prisoner finds his own name, all will be subsequently executed. Believe it or not, there is a strategy that has probability of success greater than 30%. What is this strategy?
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Colloquium
Speaker: Sigal Gottlieb (UMass Dartmouth)
Faculty Sponsor: Michael Neumann
Time: Thursday, November 13, 2008 at 4:00 pm
Place: MSB 118 (UConn - Storrs)
Abstract: When numerically solving a hyperbolic conservation law
it is important to consider the properties of the spatial
discretization combined with the time discretization.
If the problem is smooth, it is sufficient to linearize the problem and analyze the L2 stability properties of the resulting discretization. However,
if the solution is nonsmooth, stability in the L2 norm is not
sufficient. This is
because for PDEs with discontinuous solutions, the presence of
oscillations prevents
the approximation from converging uniformly. To ensure that the method
does not allow oscillations to form, we require stability of the
nonlinear system in the maximum norm
or in the TV semi-norm. Strong stability preserving (SSP) high order
time discretizations were developed
to ensure these types of nonlinear stability properties
SSP methods preserve the strong stability properties -- in any norm,
seminorm or convex functional --
of the spatial discretization coupled with first order Euler time stepping.
I will describe the development of SSP methods and the current
state-of-the-art of these methods.
|
Logic Seminar
Speaker: Paul Ellis (University of Connecticut)
Time: Monday, November 17, 2008 at 4:45 pm
Place: Exley Science Center 618 (Wesleyan University)
Abstract: It is easily seen that the conjugacy problem for the automorphism group of N is smooth. Sam Coskey, Scott Schneider and I have shown that the conjugacy problem for the automorphism group of the countable random graph is complete Borel. I will show a proof that the same is true for the automorphism group of the countable homogeneous dense linear order. Due to a curious parallel in the three proofs, we conjecture that for any countably categorical structure, the conjugacy problem for the automorphism group is either smooth (easy) or complete Borel (totally intractable). The necessary background about Borel equivalence relations will be discussed as well.
|
Algebra Seminar
Speaker: Gregg Musiker (MIT)
Time: Tuesday, November 18, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: In this talk, I will discuss examples of cluster algebras which
have graph theoretic formulas for cluster expansions. The main result is
a graph theoretical formula, as positive Laurent polynomials, for cluster
variables (with principal coefficients) arising from triangulated surfaces
without punctures.
In particular, for each cluster variable, x_gamma, we construct a
graph G_gamma, such that a weighted enumeration of the perfect
matchings of G_gamma encodes the numerator of x_gamma, decompositions of
the graphs correspond to the denominator, and we use height functions on
matchings of G_gamma to obtain coefficients.
Time permitting I will discuss other examples including cluster algebras
of classical type with bipartite seeds, and joint work with Jim Propp on
rank 2 cluster algebras of affine type.
|
Actuarial Science Seminar
Speaker: Yang Ho Choi (University of Connecticut)
Time: Tuesday, November 18, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: In the standard use of stochastic integration in pricing of option we use a model based on Brownian motion. But it restricts our flexibility. Still the fact that Brownian motion has independent increments simplifies the solution of the stochastic differential equations, such as Black-Scholes formula. Fractional Brownian motion can be compared to standard Brownian motion with the use of the Hurst parameter H. H = ½ is standard Brownian motion, and only then do we have independent increments. If H is not ½, then the mathematics involved in solving the stochastic differential equation entails new tools, and it is more tricky to convert the problem into a Partial differential equation (PDE). In this talk, we introduce the concept of fractional Brownian motion and derive fractional Black Scholes formula for a European call. We then see other topics related to fractional Brownian motion.
<Extra Information>
|
Mathematics Education
Speaker: Megan Staples and Mary Truxaw (University of Connecticut)
Time: Wednesday, November 19, 2008 at 3:00 pm
Place: Gentry 142 (UConn - Storrs)
Abstract: This talk will describe the Math ACCESS Project. This project, involving 4 Hartford Schools and 23 grade 4-9 math and special education teachers, aims to enhance student mathematical understanding by focusing on verbal and written discourse, with an emphasis on supporting higher order thinking and mathematical justification. Teachers participated in 45 hours of professional development during the summer. During the academic year, ACCESS teachers work collaboratively to design and implement higher order thinking (H.O.T.) math lessons that support student participation in explaining, justifying, and problem solving.
Research related to this project documents (a) the issues that arise for teachers as they collaborate with each other and with university researchers in an effort to develop a mathematics learning discourse in their classrooms, (b) the teacher knowledge, strategies, and approaches that were effective in resolving the identified issues, (c) the degree to which teachers do implement a mathematics learning discourse and (d) students’ participation in, and learning of, a mathematics discourse along the dimensions of meaning making, justification and the use of academic language.
|
S.I.G.M.A. Seminar
Speaker: Andrew Ursitti (University of Connecticut)
Time: Wednesday, November 19, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: In this talk we will introduce some probabilistic processes
in the complex plane which arise as the continuous scaling
limits of discrete models from statistical physics. This
Fields Medal winning work made many long standing conjectures
from physics mathematically rigorous and amenable to
computation. The talk will be mostly descriptive, in particular
nothing more than basic complex analysis will be assumed.
Comments: Free Donuts (courtesy of GSS)
|
UConn Math Club
Speaker: Dan Bochicchio (E. O. Smith High School)
Time: Wednesday, November 19, 2008 at 5:40 pm
Place: MSB 403 (UConn - Storrs)
Abstract: We will discuss different ways votes can be counted in an election and how the choice of voting method can radically change who the winner is and why. Anyone who has voted recently will find this a worthwhile talk.
Comments: Free Refreshments
Additional Comments: Email: uconnmathclub@gmail.com Facebook group: UConn Math Club
|
Colloquium
Speaker: David Rohrlich (Boston University )
Faculty Sponsor: Keith Conrad
Time: Thursday, November 20, 2008 at 4:00 pm
Place: MSB 118 (UConn - Storrs)
Abstract: In the spirit of the "What is ...?" series in the Notices of the AMS, we give a purely expository talk about modular forms, emphasizing basic definitions, examples, and motivation, and assuming no prior acquaintance with modular forms.
|
Analysis and Probability Seminar
Speaker: Jean-François Quint (Université Paris 13)
Time: Friday, November 21, 2008 at 3:00 pm
Place: MSB 319 (UConn - Storrs)
Abstract: The Pascal graph is a self-similar, planar, 3-regular graph. The spectrum
of the discrete Laplace operator on this graph has been studied by many
authors (Rammal and Toulouse, Bellissard, Kigami, Teplyaev,...). This
spectrum is strongly related to the dynamics of the polynomial
f(x)=x^2-x-3. In this talk, I will explain how one can compute the
spectral measures of certain remarkable functions on the Pascal graph. I
will deduce from this computation a spectral decomposition theorem in the
space of square-summable functions. Certain transfer operators on the
Julia set of the polynomial f will play an important role.
|
Mathematics Education
Speaker: Jeremy Teitelbaum (University of Connecticut)
Time: Monday, December 1, 2008 at 4:00 pm
Place: Gentry 142 (UConn - Storrs)
Abstract: I will present some materials I have written that
give an approach to teaching pre- and in-service teachers about the
commutative and associative laws. This is part of the more general
context
of teaching an appreciation for the "number hierarchy" (whole numbers,
integers, rational numbers....)
|
Logic Seminar
Speaker: Dan McGinn (University of Wisconsin)
Time: Monday, December 1, 2008 at 5:30 pm
Place: MSB 117 (Storrs)
Abstract: Our latest paper is part of a program to extend classical model theoretic results to the model theory of intuitionistic logic. How can one define concepts like "universal sentence", "submodel", and "model complete" in the setting of intuitionistic logic and Kripke models? In this talk, we use classical theories with nice homogeneity properties to construct certain kinds of Kripke models. The intuitionistic theories of these Kripke models are complete, prove negations of classical tautologies, and admit quantifier elimination.
This is joint work with Ben Ellison, Jonathan Fleishmann and Wim Ruitenburg.
Comments: Note -- Unusual time and location
|
Colloquium
Speaker: Boris Slepchenko (University of Connecticut Health Center)
Faculty Sponsor: Yung Sze Choi
Time: Tuesday, December 2, 2008 at 4:00 pm
Place: MSB 118 (UConn - Storrs)
Abstract: The Virtual Cell, a computational framework for modeling in cell biology, has been developed at the Center for Cell Analysis and Modeling in the Department of Cell Biology of the UConn Health Center. I will illustrate its capabilities, including the most recent ones, using examples of applications to intracellular transport and to modeling experiments on protein-membrane interactions.
Comments: UNUSUAL DATE!
|
Actuarial Science Seminar
Speaker: Benito (Jojo) Cuevo (Phoenix Life Insurance Company)
Time: Tuesday, December 2, 2008 at 4:30 pm
Place: MSB 411 (UConn - Storrs)
Abstract: Given the growing complexity of assets and insurance products coupled with the volatility of market environments in recent past, asset liability management (ALM) has become more critical than ever. This seminar gives an overview of what ALM practically means for insurers from a portfolio management and risk management perspective. The speaker discusses several ALM approaches currently used in the financial services industry as well as major ALM issues that insurers experienced during the past decade and the lessons we learned from them. This one-hour discussion also provides an overview of asset allocation for the insurance industry.
<Extra Information>
|
S.I.G.M.A. Seminar
Speaker: Stephen Miller (University of Connecticut)
Time: Wednesday, December 3, 2008 at 4:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: We will discuss equivalences between cohomology theories, spectra and linear functors. We will introduce some basic tools of homotopy theory, quote Brown Representability and mention functor calculus. This talk should be accessible to graduate students.
Comments: Free Donuts (courtesy of GSS)
|
Algebra Seminar
Speaker: Russell Prime (University of Connecticut)
Time: Tuesday, December 9, 2008 at 3:30 pm
Place: MSB 118 (UConn - Storrs)
Abstract: Gauss posed a problem about the average value of class numbers
of quadratic number fields, and by Dirichlet's class number formula, this
can be generalized to a question about the average value of L(s,chi) over
quadratic characters chi. In part I of the talk, we will discuss the
results that address this problem, and introduce the analogous ideas in
the function field setting. In part II of the talk, I will talk about my
current work in averaging L-functions over function fields, and discuss a
proof of a result concerning the average of L(s,chi) for quadratic chi
with irreducible modulus.
|
|
|
University of Connecticut | 
