Home > Events > SMAC Talks > Stochastic Modeling and Computational Statistics, Spring 2013

Stochastic Modeling and Computational Statistics, Spring 2013

Main Content


Guidelines:

  1. 40 minutes for each talk + 20 minutes for discussion
  2. informal style: chalk and blackboard talks are welcome
  3. the talk should be accessible to all grad students who have completed 1 year of the program
  4. interruptions during the talk are welcome but they should only be for clarifications; other questions are to be left to discussions
  5. unpublished work may not be shared/discussed outside the group without the permission of the speaker/author

 

 

DateSpeakerTopic
Jan 10 Vishesh Karwa An introduction to propensity score methods
Jan 17 No meeting
Jan 24 Chia-Jung Chang (PSU Industrial Engineering) Representation/quantification of nano-particle dispersion using non-homogeneous Poisson process
Jan 31 Jason Morton (Math/Stat) Graphical hypermodels: a technique for interpolating among Bayesian network models with differing architecture
Feb 7 Dennis Lin Dimensional Analysis and Statistics
Feb 14 No meeting
Feb 21 Michael Schweinberger Second-generation exponential-family models of networks: Scaling up
Feb 28 Youngjoo Cho A Weighted Estimator of Accelerated Failure Time Model under Presence of Dependent Censoring
Mar 7 No meeting due to spring break
Mar 14 Boaz Nadler (Weizmann Institute) Diffusion maps, a generalization of principal components for nonlinear (non-Gaussian) data
Mar 21 Yu Zhang Infinite-state HMMs for population sequencing data analysis
Mar 28 Don Richards The Affinely Invariant Distance Correlation
Apr 4 Uday Shanbhag (PSU Industrial Engineering/Operations Research) Stochastic approximation schemes: regularization, smoothing, and adaptive steplengths
Apr 11 Michael Kuhn (Astronomy) Spatial point processes to study clusters of young stars
Apr 18 Won Chang A composite likelihood approach to computer model calibration
Apr 25 Murali Haran Tutorial/overview of Monte Carlo methods for spatial generalized linear mixed models