Analysis of Monte Carlo estimators for parametric sensitivities in stochastic chemical kinetics
46 mins 11 secs,
176.64 MB,
iPod Video
480x270,
29.97 fps,
44100 Hz,
522.19 kbits/sec
Share this media item:
Embed this media item:
Embed this media item:
About this item
Description: |
Rathinam, M (University of Maryland, Baltimore County)
Tuesday 5th April 2016 - 09:45 to 10:30 |
---|
Created: | 2016-04-06 10:23 |
---|---|
Collection: | Stochastic Dynamical Systems in Biology: Numerical Methods and Applications |
Publisher: | Isaac Newton Institute |
Copyright: | Rathinam, M |
Language: | eng (English) |
Distribution: | World (downloadable) |
Explicit content: | No |
Aspect Ratio: | 16:9 |
Screencast: | No |
Bumper: | UCS Default |
Trailer: | UCS Default |
Abstract: | Co-author: Ting Wang (University of Delaware)
We provide an overview of some of the major Monte Carlo approaches for parametric sensitivities in stochastic chemical systems. The efficiency of a Monte Carlo approach depends in part on the variance of the estimator. It has been numerically observed that in several examples, that the finite difference (FD) and the (regularized) pathwise differentiation (RPD) methods tend to have lower variance than the Girsanov Tranformation (GT) estimator while the latter has the advantage of being unbiased. We present a theoretical explanation in terms of system volume asymptotics for the larger variance of the GT approach when compared to the FD methods. We also present an analysis of efficiency of the FD and GT methods in terms of desired error and system volume. |
---|
Available Formats
Format | Quality | Bitrate | Size | |||
---|---|---|---|---|---|---|
MPEG-4 Video | 640x360 | 1.94 Mbits/sec | 672.43 MB | View | Download | |
WebM | 640x360 | 570.6 kbits/sec | 193.08 MB | View | Download | |
iPod Video * | 480x270 | 522.19 kbits/sec | 176.64 MB | View | Download | |
MP3 | 44100 Hz | 249.73 kbits/sec | 84.57 MB | Listen | Download | |
Auto | (Allows browser to choose a format it supports) |