Please click here for upcoming seminar information as well as other events going on in ICME.
MONDAY, SEPTEMBER 26, 2016: CME 500: Departmental Seminar
There will be no seminar this week.
TUESDAY, SEPTEMBER 27, 2016: CME 300: First Year Seminar Series
This week, ICME Director, Margot Gerritsen, will be having an informal meeting with first year students to discuss goals and upcoming events at ICME.
Location: Y2E2-101 Time: 12:30-1:20 p.m.
THURSDAY, SEPTEMBER 29, 2016: CME 242: Mathematical and Computational Finance Seminar
Speaker: Ronald Gallant, Pennsylvania State University
Title: Bayesian Inference using the EMM Objective Function: With Application to the Dominant Generalized Blumenthal-Getoor Index of an Ito Semimartingale
Abstract: We use the Gallant and Tauchen (1996) efficient method of moments (EMM) method modified to accomodate Bayesian inference to estimate the generalized Blumenthal-Getoor index of S&P 500 index options and futures over the period January 3, 2007, to March 22, 2011 using data based on Andersen et al. (2013, 2015). The options extend over a wide range of moneyness, including deep out of the money puts. As explained by high frequency financial theory, the seminonparametric density estimates of pre-averaged 5-minute returns on options are strongly consistent with a pure jump model with β-stable jumps, 1 < β < 2. When fitted via EMM with a prior that imposes support conditions implied by theory, the β-stable gives estimates close to 1 (Cauchy) for deep out of the money options and estimates around 1.80 for at the money. The pattern suggests a far more important than previously thought role for jump-like behavior and diminished, if not absent, diffusive component. Our results greatly sharpen those of recent studies that reveal the relevance of pure jump models with many very small or moderate jumps, plus rare jumps, instead of the classic continuous model comprised of a continuous diffusion plus rare jumps.
Location: 200-205
Time: 4:30-5:50 p.m.
THURSDAY, SEPTEMBER 29, 2016: CME 510: Linear Algebra and Optimization Seminar
Speakers: Ding Ma and Michael Saunders, MS&E and ICME at Stanford University
Title: The DQQ procedure for multiscale optimization
Abstract: Constrained optimization solvers typically scale the constraints, solve the scaled problem, then unscale. With multiscale problems and a conventional double-precision solver, the unscaled solution may not satisfy the constraints well. Our DQQ procedure continues with a quad-precision version of the solver to obtain accurate solutions efficiently.
A prime application is to metabolic networks in systems biology. Keywords are flux balance analysis (FBA) and genome-scale modeling of Metabolism and macromolecular Expression (ME models). DQQ typically achieves at least 20 digits of precision.
For smooth functions with few variables, Quad improves the performance of quasi-Newton optimization with finite-difference gradients. We illustrate with IMSPE-optimal design of computer experiments (joint work with Selden Crary).
Our mailing address is:
Institute for Computational and Mathematical Engineering (ICME)
Huang Engineering Center
475 Via Ortega, Suite B060
Stanford, CA 94305
If you would like to unsubscribe from the ICME seminar digest, please email us at icme contact@stanford.eduto be completely removed from our mailing list. Thank you.