BEGIN:VCALENDAR
PRODID:-//PSU Mathematics Department//Seminar iCalendar Generator//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Probability and Financial Mathematics Seminar
X-WR-TIMEZONE:America/New_York
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20140822T153500
DTEND;TZID=America/New_York:20140822T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22054
SUMMARY:Probability and Financial Mathematics Seminar - TBA
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
TBA
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20140829T153500
DTEND;TZID=America/New_York:20140829T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22055
SUMMARY:Probability and Financial Mathematics Seminar - TBA
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
TBA
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20140905T153500
DTEND;TZID=America/New_York:20140905T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22056
SUMMARY:Probability and Financial Mathematics Seminar - Weighted entropy
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
Weighted entropy\nSpeaker: Yuri Suhov\, Penn State/University of Cambridge
\, UK\nAbstract: The entropy $h(X)=-\\sum_ip_i\\log\\\,p_i$ measures an ex
pected amount of information/uncertainty related to a random variable $X$
taking values $i$ with probabilities $p_i$. The weighted entropy\, $h^{\\r
m w}_\\phi(X)$\, is defined as $-\\sum_i\\phi (i)p_i\\log\\\,p_i$ where $
\\phi (i)\\geq 0$ is a weight function representing `utilities' of differe
nt values $i$ which we want to take into account. As in the case of a stan
dard entropy\, one can introduce conditional and relative weighted entropi
es\; weighted differential entropies can also be defined. In this talk\, I
will discuss a recent progress in studying weighted entropies and their p
ossible use in various areas.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20140912T153500
DTEND;TZID=America/New_York:20140912T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22057
SUMMARY:Probability and Financial Mathematics Seminar - No meeting
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
No meeting
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20140919T153500
DTEND;TZID=America/New_York:20140919T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22058
SUMMARY:Probability and Financial Mathematics Seminar - A nonconventional l
ocal limit theorem
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
A nonconventional local limit theorem\nSpeaker: Yuri Kifer\, Hebrew Univer
sity\nAbstract: Local limit theorems have their origin in the classical De
Moivre– Laplace theorem and they study the asymptotic behavior as N →
∞ of probabilities having the form P {S_N = k} where S_N = \\sum^N_{n=1
} F (ξ_n ) is a sum of an integer valued function F taken on i.i.d. or Ma
rkov dependent sequence of random variables {ξ_j}. Corresponding results
for lattice valued and general functions F were obtained\, as well. We ext
end here this type of results to nonconventional sums of the form S_N =
\\sum^N_{n=1} F (ξ_n \, ξ_{2n} \, ...\, ξ_{ln} ) and corresponding vers
ions of such results can be obtained for some dynamical systems\, as well.
This continues the recent line of research studying various limit theorem
s for such expressions.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20140926T153500
DTEND;TZID=America/New_York:20140926T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22059
SUMMARY:Probability and Financial Mathematics Seminar - The Widom--Rowlinso
n model and the phenomenon of allelopathy
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
The Widom--Rowlinson model and the phenomenon of allelopathy\nSpeaker: Yur
i Suhov\, Penn State/University of Cambridge\, UK\nAbstract: The Widom--Ro
wlinson (WR) model was proposed in early 1970s in Chemical Physics\, to ex
plain various phenomena at the molecular level. There are several types of
particles which repel one another when they belong to different types and
have no influence upon each other if they belong to the same type (these
assumptions can be made more general). This resembles an allelopathic phen
omenon observed in biology where a given species prevents other species fr
om occupying the space nearby by using a variety of means (poisoning soil
or water\, encouraging parasites harmful to other species but harmless to
themselves\, etc.). There is also a quantum version of the model.\n\nThe W
R model became popular in various disciplines. An interesting question is
about phase transitions: if the overall particle density is low\, there is
one equilibrium (Gibbs) distribution resembling Poisson. However\, if the
density is high\, there may be one or several distributions where a parti
cular type will dominate (occupy an overwhelming proportion of the space).
This is determined by the collection of hard-core repulsion diameters. Th
e talk will focus on new results on this question and emerging application
s.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141003T153500
DTEND;TZID=America/New_York:20141003T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22060
SUMMARY:Probability and Financial Mathematics Seminar - Ergodicity of avala
nche transformations
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
Ergodicity of avalanche transformations\nSpeaker: Manfred Denker\, PSU\nAb
stract: An avalanche transformation is a product transformation followed b
y an avalanche dynamics. The talk will provide a precise definition. I w
ill discuss the question when such a transformation is ergodic\, besides o
ther questions like topological transitivity and central limit theorems.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141010T153500
DTEND;TZID=America/New_York:20141010T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22061
SUMMARY:Probability and Financial Mathematics Seminar - On infinitely divis
ible semimartingales
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
On infinitely divisible semimartingales\nSpeaker: Jan Rosinski\, Universit
y of Tennessee\nAbstract: Semimartingales play a fundamental role in stoch
astic analysis and mathematical finance. Concerning the latter\, the disc
ounted asset price process must be a semimartingale in order to preclude
arbitrage opportunities. The question whether a given process with long me
mory\, possible jumps and/or heavy tails is a semimartingale is also of im
portance in stochastic modeling\, where such processes are used as a driv
ing random motion for stochastic differential equations. We consider this
question in the context of infinitely divisible processes\, which include
fractional processes\, moving averages\, and Ornstein-Uhlenbeck processe
s driven by stable\, multi-stable\, and tempered stable L\\'evy processes
\, and their mixtures. We show that the problem when any such process is a
semimartingale can often be reduced to a path property\, when a certain a
ssociated infinitely divisible process is of finite variation. This gives
the key to fully characterize the semimartingale property for many process
es of interest\, including processes mentioned above.\n\nThis talk is base
d on a joint work with Andreas Basse-O'Connor of Aarhus University.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141017T153500
DTEND;TZID=America/New_York:20141017T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22062
SUMMARY:Probability and Financial Mathematics Seminar - Heavy-traffic limit
s for a fork-join network in the Halfin-Whitt regime
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
Heavy-traffic limits for a fork-join network in the Halfin-Whitt regime\nS
peaker: Hongyuan Lu\, PSU\, Industr. & Manufact. Engineering\nAbstract: We
study a fork-join network with a single class of jobs\, which are forked
into a fixed number of parallel tasks upon arrival to be processed at the
corresponding parallel service stations. After service completion\, each t
ask will join a buffer associated with the service station waiting for syn
chronization\, called ``unsynchronized queue". The synchronization rule re
quires that all tasks from the same job must be completed\, referred to ``
non-exchangeable synchronization". Once synchronized\, jobs will leave the
system immediately. Service times of the associated parallel tasks of eac
h job can be correlated and form a sequence of i.i.d. random vectors with
a general continuous joint distribution function. Each service station has
multiple statistically identical parallel servers. We consider the system
in the Halfin-Whitt (Quality-and-Efficiency-Driven\, QED) regime\, in whi
ch the arrival rate of jobs and the number of servers in each station get
large appropriately so that all service stations become critically loaded
asymptotically.\n\n\nWe develop a new method to study the joint dynamics o
f the service processes and the unsynchronized queueing processes at all s
tations and the synchronized process. The waiting processes for synchroniz
ation after service depend on the service dynamics at all service stations
\, and thus are extremely difficult to analyze exactly. The main mathemati
cal challenge lies in the resequencing of arrival orders after service com
pletion at each station. We represent the dynamics of all the aforemention
ed processes via a multiparameter sequential empirical process driven by t
he service vectors of the parallel tasks. We show a functional law large n
umber (FLLN) and a functional central limit theorem (FCLT) for these proce
sses. Both the service and unsynchronized queueing processes in the limit
can be characterized as unique solutions to the associated integral convol
ution equations driven by the arrival limit process and a generalized mult
iparameter Kiefer process driven by the service vectors.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141024T153500
DTEND;TZID=America/New_York:20141024T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22063
SUMMARY:Probability and Financial Mathematics Seminar - When does a mixture
of products contain a product of mixtures?
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
When does a mixture of products contain a product of mixtures?\nSpeaker: J
ason Morton\, PSU\nAbstract: We derive relations between theoretical prope
rties of restricted Boltzmann machines (RBMs)\, popular machine learning m
odels which form the building blocks of deep learning models\, and several
natural notions from discrete mathematics and convex geometry. We give im
plications and equivalences relating RBM-representable probability distrib
utions\, perfectly reconstructible inputs\, Hamming modes\, zonotopes and
zonosets\, point configurations in hyperplane arrangements\, linear thresh
old codes\, and multi-covering numbers of hypercubes. As a motivating appl
ication\, we prove results on the relative representational power of mixtu
res of product distributions and products of mixtures of pairs of product
distributions (RBMs) that formally justify widely held intuitions about di
stributed representations. In particular\, we show that an exponentially l
arger mixture of products\, requiring an exponentially larger number of pa
rameters\, is required to represent the probability distributions represen
ted as products of mixtures.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141031T153500
DTEND;TZID=America/New_York:20141031T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22064
SUMMARY:Probability and Financial Mathematics Seminar - On time inhomogeneo
us branching Brownian motion
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
On time inhomogeneous branching Brownian motion\nSpeaker: Alexei Novikov\,
PSU\nAbstract: A binary branching Brownian motion is a continuous-time Ma
rkov branching process that is constructed as follows: start with a single
particle which performs a standard Brownian motion x(t) with x(0) = 0 and
continues for an exponentially distributed holding time T\, independent o
f x. At time T\, the particle splits independently of x and T into 2 offsp
ring with probability p. We discuss what happens if the variance of the Br
ownian motion depends on time.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141107T153500
DTEND;TZID=America/New_York:20141107T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22065
SUMMARY:Probability and Financial Mathematics Seminar - Limit distribution
of return times to shrinking targets
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
Limit distribution of return times to shrinking targets\nSpeaker: Xuan Zh
ang\, PSU\nAbstract: We revisit the problem of finding limit exponential l
aws for return times to shrinking targets and study a more general problem
of finding limit distribution for stopping times when partial sums of bou
nded positive functions exceed a threshold. We give a sufficient condition
for the limit distribution of these stopping times to exist and discuss s
ome examples.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141114T153500
DTEND;TZID=America/New_York:20141114T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22066
SUMMARY:Probability and Financial Mathematics Seminar - Convergence of stoc
hastic processes to Brownian local time
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
Convergence of stochastic processes to Brownian local time\nSpeaker: Xiaof
eng Zheng\, PSU\nAbstract: I will introduce some results about the converg
ence of various sequences of processes describing the behavior of Brownia
n trajectories near the level x at the instant t to Brownian local time.
We are also interested in the question whether local time of random walks
can be convergent to Brownian local time almost surely. Peter Morters and
Yuval Peres gave a positive answer to the subsequence of local time of ra
ndom walks case by embedding method\, which will also be discussed.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141121T153500
DTEND;TZID=America/New_York:20141121T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22067
SUMMARY:Probability and Financial Mathematics Seminar - Characterization of
stationary distributions of reflected diffusions
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
Characterization of stationary distributions of reflected diffusions\nSpea
ker: Weining Kang\, University of Maryland\nAbstract: In this talk I will
focus on a characterization of stationary distributions of reflected diffu
sions that arise as approximations to stochastic systems commonly from tel
ecommunications and manufacturing\, etc. I will describe the extended Skor
okhod problem formulation and the submartingale problem formulation of the
class of reflected diffusions and explore their connections. Also I will
discuss how their connections can be used to give a characterization of th
eir stationary distributions and how this characterization can be used to
compute those stationary distributions. Some additional applications of th
is characterization will also be discussed.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141128T153500
DTEND;TZID=America/New_York:20141128T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22068
SUMMARY:Probability and Financial Mathematics Seminar - No meeting
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
No meeting
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141205T153500
DTEND;TZID=America/New_York:20141205T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22069
SUMMARY:Probability and Financial Mathematics Seminar - A distribution for
avalanche sizes in neural dynamics
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
A distribution for avalanche sizes in neural dynamics\nSpeaker: Anirban Da
s\, PSU\nAbstract: In 2002 Eurich\, Herrmann and Ernst proposed a distribu
tion for the size of avalanches in complete networks of neurons without le
aking (Phys. Rev. E\, 2002). The talk will present some new additional pro
perties for this and similar distributions\, in particular\, a formula for
the variance and a complete derivation of Wenbo Li's formula for expectat
ions.
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141212T153500
DTEND;TZID=America/New_York:20141212T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22070
SUMMARY:Probability and Financial Mathematics Seminar - No meeting
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
No meeting
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20141219T153500
DTEND;TZID=America/New_York:20141219T163500
LOCATION:MB106
URL:http://www.math.psu.edu/seminars/meeting.php?id=22071
SUMMARY:Probability and Financial Mathematics Seminar - TBA
DESCRIPTION:Seminar: Probability and Financial Mathematics Seminar\nTitle:
TBA
END:VEVENT
END:VCALENDAR