Exploring ODE model uncertainty via diffusions, with application to physiological processes

Add to your list(s) Download to your calendar using vCal

• Dr Kostas Kalogeropoulos, London School of Economics
• Wednesday 19 May 2010, 10:00-11:00
• LR12, Engineering, Department of.

If you have a question about this talk, please contact Rachel Fogg.

Ordinary differential equations models are extensively used to describe various continuous time phenomena. Although successful in capturing certain aspects of the underlying mechanism, the need for further refinement is becoming increasing apparent. In many applications, including our motivating example from physiological processes, this may be attributed to the inherent system noise. In our approach the system noise is explicitly modelled and disentangled from other sources of error. This is achieved by incorporating a diffusive component while retaining the mean infinitesimal behavior of the system. In PK/PD applications, the problem of inference on the non linear diffusion parameters is further complicated by the presence of measurement error, individual variability, imbalanced designs and so forth. We present a general inference framework through data augmentation which we illustrate through simulated and real PK/PD data.

This talk is part of the Signal Processing and Communications Lab Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge University Engineering Department Talks
• Cambridge talks
• Centre for Smart Infrastructure & Construction
• Chris Davis' list
• Computational Continuum Mechanics Group Seminars
• Featured lists
• Information Engineering Division seminar list
• Interested Talks
• LR12, Engineering, Department of
• School of Technology
• Signal Processing and Communications Lab Seminars
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.