DISSERTATION DEFENSE ANNOUNCEMENT

DISSERTATION DEFENSE ANNOUNCEMENT
TITLE:                      A Continuous-Time Model for the Valuation of Mortgage-Backed Securities
SPEAKER:                Stephen M. Mansour, PhD Candidate
                                    Department of Industrial and Systems Engineering

DATE:                        Friday, December 9, 2011 from 3:30 – 5:30 pm 
LOCATION:             Room 451 Mohler, 200 W. Packer Avenue

ABSTRACT:  A mortgage model consists of three basic parts:  the amortization model which examines the mortgage cash flows, the interest rate model which affects the mortgage price, and the prepayment model which measures the rates of mortgage termination when a property is sold, refinanced or foreclosed.  A technique known as eigenfunction expansion has proven to be useful in pricing continuous-time mortgages.   

The first part of this presentation involves generalizing the existing interest rate Cox-Ingersoll-Ross model and including as an alternative the simpler Vasicek model and then comparing the results obtained by these methods.  We also refine the relationship between interest rates and prepayments to reflect empirical data more accurately, particularly in low-interest rate scenarios by expanding the existing single-threshold prepayment model to include a secondary prepayment threshold.   

The second problem expands the existing continuous prepayment model to include mortgage defaults.   We use the default model to examine the price sensitivity of mortgages to loss severity and foreclosure rates.  We also examine two practical applications of this model:  accounting for wider spreads between mortgage yields and treasury yields during periods of economic stress, and estimating the value of the mortgage guarantee that government agencies such as Ginnie Mae provide to investors of mortgage-backed securities.   

BIOGRAPHY:  Stephen M. Mansour is a PhD candidate in the Department of Industrial and Systems Engineering at Lehigh University.   He received a Bachelor’s Degree in Mathematics with Distinction in General Scholarship from the University of California at Berkeley in 1981 where he graduated Phi Beta Kappa.  He worked at IBM East Fishkill, New York from 1982 to 1994 as an APL programmer. While at IBM he received a Division Award for “Outstanding Team Leadership During the Development of the ALORS2 Data Base”. During his tenure at IBM, he also received a Master’s Degree in Operations Research and Applied Statistics from Union College in 1992.   He worked at Check-Free Corporation in Jersey City, New Jersey from 1994-1996 where he developed a billing system for portfolio managers, and at The Carlisle Group in Scranton, Pennsylvania from 1996-2008 where he developed an APL-based pricing and portfolio optimization system for use by mortgage companies.   He is currently teaching statistics at the University of Scranton and at Penn State Worthington Scranton.

Dan Scansaroli's Ph.D. defense 12-1-11

DISSERTATION DEFENSE ANNOUNCEMENT

TITLE:                    Stochastic Modeling with Temporally Dependent Gaussian Processes: Applications to Financial Engineering, Pricing and Risk Management

SPEAKER:            Daniel J. Scansaroli, PhD Candidate Department of Industrial and Systems Engineering

DATE:                   Thursday, December 1, 2011 from 3:00 – 5:00 pm

LOCATION:         Room 375, Mohler Laboratory, 200 W. Packer Avenue

ABSTRACT: This thesis studies two classes of the most often applied temporally dependent Gaussian processes. Computationally efficient and accurate techniques are
developed for modeling and parameter estimation with the goal of improving decision making, risk management, pricing and hedging in finance.

We first focus on the widely used fractional Brownian motion (fBm) processes. We explore the advantages and disadvantages of modeling with the processes and present
new consistent estimators of the Hurst index for a Weiner type fBm process. Simulation studies demonstrate that the new estimators are highly competitive to leading estimators
in accuracy, especially on small data sets, and much more time efficient. This makes the estimators ideal for fast paced financial markets, which is demonstrated on a variety of indices.
We conclude our study by demonstrating that the dependency structure of fBm may explain the term structure of volatility commonly observed in practice.

The second part of this presentation focuses on Gaussian Markov (GM) processes. GM processes allow for a wide range of properties including long or short-range dependence,
 non-stationarity, and heteroscedasticity. We prove that the quadratic variation leads to a new, computationally efficient, consistent estimator of a model’s diffusion parameter.
Consistency is proven on a finite time interval, making it well suited for real world applications. This contrasts with existing MLE methods that require an infinite time horizon for
consistency. The convergence rate and confidence interval bounds for the estimator are also obtained. We demonstrate how the quadratic variation changes Option Pricing Theory
and extend the Black-Scholes formula for general continuous sample path GM processes.

The accuracy of diffusion parameter estimation techniques is demonstrated by simulating an Ornstein-Uhlenbeck process and applying the quadratic variation and Maximum Likelihood
estimators. We use the Likelihood function to express all model parameter estimators in closed-form, eliminating the need for estimation through three dimensional numerical optimization methods.

The final part of the presentation addresses the pricing of American style derivatives through the discretization of any continuous path GM processes into a recombining n-period binomial tree. The
Central Limit Theorem for Stochastic Processes is used to prove that the tree converges to its continuous path GM process. We apply our method to create a tree for the Vasicek interest rate model
and price an American put option.

BIOGRAPHY: Daniel Jonathan Scansaroli is a Ph.D. candidate in the Department of Industrial and Systems Engineering at Lehigh University. Educated at Lehigh for both undergraduate and graduate
degrees, he received a Bachelor of Science in Mechanical Engineering in 2005. Awarded Lehigh University's Presidential Scholarship, he received the degree of Master of Science in
Applied Mathematics in May 2006 and a Master of Science in Management Science in January 2009. Currently, Dan is employed in asset management as a quantitative analyst for
Lehigh University's endowment office.

Coopr 3.1 Release

It is pleased to announce the release of Coopr 3.1 (3.1.5325). Coopr
is a collection of Python software packages that supports a diverse
set of optimization capabilities for formulating and analyzing
optimization models.

The following are highlights of this release:

- Solvers
   * Interfaces for !OpenOpt solvers
   * Many solver interface improvements
   * A solver checker to validate solver interfaces
   * Improved support for SOS constraints (cplex, gurobi)
   * PH supports nonlinear models
   * PH-specific solver servers

- Modeling
   * Comprehensive rework of blocks and connectors for modular modeling
   * New !VarList component
   * Added comprehensive support for set expressions

- Usability enhancements
   * New 'coopr' command has subcommands that consolidate Coopr scripting
   * Added support to connect to databases with ODBC
   * Made JSON the default results format

- Other
   * Efficiency improvements in model generation, memory, runtime, etc.
   * Preliminary support for black-box applications
   * Deprecated modeling syntax in Coopr 3.0 is no longer legal

See https://software.sandia.gov/trac/coopr/wiki/GettingStarted for
instructions for getting started with Coopr.  Installers are available
for MS Windows and Unix operating systems to simplify the installation
of Coopr packages along with the third-party Python packages that they
depend on.  These installers can also automatically install extension
packages from Coin Bazaar.

Enjoy!

Householder Fellowship at the Oak Ridge National Laboratory

Householder Fellowship

Purpose The Computer Science and Mathematics (CSM) Division at the Oak Ridge

National Laboratory (ORNL) invites outstanding candidates to apply for

the Alston S. Householder Fellowship in Scientific Computing.

Description

The Fellowship honors Dr. Alston S. Householder, founding director of the

Mathematics Division (now CSM Division) at ORNL and recognizes his

seminal research contributions to the fields of numerical analysis and

scientific computing. Funding for the Householder Fellowship comes from

the Computational Mathematics Project, which is supported by the Office

of Mathematical, Information, and Computational Sciences of the U.S.

Department of Energy (http://www.science.doe.gov/ascr).

Additional information about the math group at ORNL can be found at

http://www.csm.ornl.gov/comp_math/.

The purpose of the Householder Fellowship is to promote innovative

research in scientific computing on advanced computer architectures and

to facilitate technology transfer from the laboratory research

environment to industry and academia through advanced training of new

computational scientists. The applied mathematics research efforts

provide the fundamental mathematical methods and algorithms needed to

model complex physical, chemical, and biological systems. The computer

science research efforts enable scientists to efficiently implement

these models on the highest performance computers available and to

store, manage, analyze, and visualize the massive amounts of data that

result. Networking research provides the techniques to link the data

producers, e.g., supercomputers and large experimental facilities, with

the data consumers, i.e., scientists who need the data.

Qualifications Required The position requires a Ph.D. in computer science, mathematics, or

statistics. Candidates nearing completion of the Ph.D.but no more than

three years beyond completion can be considered. The successful

candidate will have a strong background and interest in multi-scale

methods for scientific computing. Principal research areas include:

- boundary element method

- dense matrix computations

- direct methods for sparse matrix computations

- iterative methods for linear systems

- algorithms for solving differential equations

- large eigenvalue computations

- computational geometry and mesh generation

* A security clearance is not required for this position

SELECTION: Finalists for the Fellowship will be invited to visit ORNL to present a seminar and visit the area. The selected Fellow must be available to begin the appointment during calendar year 2012.Appointments are for one year with the option to renew for a second year. Each Householder Fellowship is a staff-level appointment that provides access to state-of-the-art computational facilities(high-performance workstations and parallel architectures), and collaborative research opportunities in active research programs in

To Apply

Please visit http://jobs.ornl.gov to officially apply to this Fellowship. For additional questions please contact Kate Carter at carterka@ornl.gov or Ed D'Azevedo dazevedoef@ornl.gov.

Announcing the winner of the 2011 COIN-OR INFORMS Cup

The submission "OpenSolver: Open Source Optimisation for Excel using COIN-OR", by Andrew Mason and Iain Dunning, has been selected as the winner of the 2011 edition of the COIN-OR INFORMS Cup. OpenSolver is an "Open Source linear and integer optimizer for Microsoft Excel. OpenSolver is an Excel VBA add-in that extends Excel’s built-in Solver with a more powerful Linear Programming solver." (from http://opensolver.org)

All entrants and their supporters are welcome to join in the celebration and regale (rile) the prize winners.

Date: Sunday, November 13
Time: 8pm-10pm

Location: The Fox and Hound
330 North Tryon St.
Charlotte, NC 28202
(Directions: http://tinyurl.com/75zhm7k)

The celebration is sponsored by IBM.

Thanks to all those who submitted to the COIN-OR INFORMS Cup.

The COIN-OR INFORMS Cup committee:

Pietro Belotti
Matthew Galati
R. Kipp Martin
Stefan Vigerske

Position in Computational Stochastic Programming at Sandia National Laboratories

The Discrete Mathematical and Complex Systems Department of Sandia National Laboratories in Albuquerque, New Mexico, is soliciting CVs  for consideration regarding a position in computational stochastic programming, with specific application to day-ahead planning for the electrical grid. 

The position is anticipated to be open for 2 years, with the possibility of extension. Requirements for the position include:

• Significant practical experience in group coding projects, using one of: C++, C, or Python.
• A working understanding of mathematical programming, including core modeling tools and algorithms.
• Experience formulating, solving, and analyzing stochastic programs.
• A minimum of an MS (PhD preferred, but not required) in Computer Science, Operations Research, or a related field.

Additional skills desired include:

• Experience with electrical grid applications.
• Experience with high-performance computing, specifically large-scale distributed-memory clusters.

If you meet these requirements, please e-mail your CV to: jwatson@sandia.gov (Dr. Jean-Paul Watson) for potential consideration.