1 / 13

Modeling and Estimating Granger Causality between Signals

Modeling and Estimating Granger Causality between Signals. Syed Ashrafulla October 11, 2013. Granger Causality …. models for testing. restricted:. unrestricted:. Hypothesis:. rejection implies Granger causality. assume Gaussianity  the signals are defined by their mean & variance.

malise
Télécharger la présentation

Modeling and Estimating Granger Causality between Signals

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Modeling and Estimating Granger Causality between Signals Syed Ashrafulla October 11, 2013

  2. Granger Causality … models for testing restricted: unrestricted: Hypothesis: rejection implies Granger causality assume Gaussianity the signals are defined by their mean & variance

  3. Granger Causality … Ebert-Uphoff2012, Geophys Research Lett Michalareas2012, HBM Geweke1982, J Amer Stat Assoc GNP deflator vs M1 geopotential height economics climatology genetics neuroscience ecology expression profile of HeLa kniematics & music Fujita2007, BMC Systems Biology D’Ausillo2012, PLoS ONE

  4. … in Mean: Canonical Granger Causality Granger causality Ashrafulla2013, NeuroImage

  5. … in Mean: Canonical Granger Causality how big is this number? Ashrafulla2012, Proc IEEE ISBI  most causal signals between two sets

  6. … in Mean: Canonical Granger Causality application: visuomotor processing Bressler1993, Nature Ashrafulla2013, NeuroImage

  7. … in Variance: ARCH & ADMM signal-dependent variation Papiez2013, Energy Econ

  8. … in Variance: ARCH & ADMM conditional heteroscedasticity: how big are these numbers? model:

  9. … in Variance: ARCH & ADMM maximum likelihood: penalized ML fixed-point iteration least-squares dual ascent ADMM: Ashrafulla2013, Asilmoar SSC

  10. … in Variance: ARCH & ADMM univariate model ADMM is faster than current methods to estimate ARCH

  11. … in Variance: ARCH & ADMM bivariate model ADMM is faster than current methods to estimate ARCH

  12. Conclusions causality in mean: a new model causality in variance: a faster method ADMM: Granger causality Ashrafulla2012, Proc IEEE ISBI Ashrafulla2013, NeuroImage Ashrafulla2013, Asilmoar SSC

  13. Journal Papers • Ashrafulla, S., Haldar, J. P., Joshi, A. A. & Leahy, R. M. (2013). Canonical Granger causality between regions of interest.NeuroImage, 83, 189–199. • Ashrafulla, S., Pantazis, D., Mosher, J., Hämäläinen, M. & Leahy, R. M. (2013). Multicenter consistency of localization of MEG. (in preparation) • Niv, S., Ashrafulla, S., Tuvblad, C., Joshi, A.A., Raine, A., Leahy, R. M. & Baker, L.A. (2013). Childhood EEG frontal alpha power as a predictor of adolescent antisocial behavior: a twin heritability study. (submitted) • Niv, S., Ashrafulla, S., Tuvblad, C., Joshi, A.A., Raine, A., Leahy, R. M. & Baker, L.A. (2013). Relationships of alpha, beta, and theta EEG spectral properties with aggressive and nonaggressive ASB in children and adolescents. (submitted) • Joshi, A.A., Ashrafulla, S., Damasio, H., Shattuck, D. W. & Leahy, R. M. (2013). Automated generation, representation and analysis of sulcal curves on human cortex. (submitted) • Conference Submissions • Ashrafulla, S., Mosher, J. C. & Leahy, R. M. (2013). Causality in variance in electrophysiological data Using the GARCH model, Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, USA. • Ashrafulla, S., Haldar, J. P., Joshi, A. A. & Leahy, R. M. (2012). Canonical Granger causality. IntConfBiomag. Paris, France. • Ashrafulla, S., Haldar, J. P., Joshi, A. A. & Leahy, R. M. (2012). Canonical Granger causality applied to functional brain data. Proc IEEE ISBI (pp. 1751–1754). Barcelona, Spain: IEEE. • Ashrafulla, S., Pantazis, D., Mosher, J. C., Hämäläinen, M., Liu, B. J. & Leahy, R. M. (2011). Viability of sharing MEG data using minimum-norm imaging. (W. W. Boonn & B. J. Liu, Eds.), 7967, 79670F–79670F–8. • Ashrafulla, S., Tadel, F., Baillet, S., Liu, B. J., Mosher, J. C., Khan, S., Lefevre, J., Huang, H. K. & Leahy, RM (2010). An imaging informatics tool for visualization of cortical flow in epilepsy via EEG. RSNA. • Ashrafulla, S., Pantazis, D. & Leahy, R. M. (2010). Evaluation of cross-site consistency of MEG data. IntConfBiomag, 4, 90089. • Aydore, S., Ashrafulla, S., Joshi, A. A. & Leahy, R. M. (2013). A measure of connectivity in the presence of crosstalk, Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, USA. • Joshi, A. A., Ashrafulla, S., Shattuck, D. W., Damasio, H. & Leahy, R. M. (2013). Cortical shape analysis using the anisotropic Global Point Signature, MICCAI 16, MFCA workshop, Nagoya, Japan. • Dery, S., Aydore, S., Ashrafulla, S., Florin, E., Bock, E., Leahy, R. M., Baillet, S. & Tadel, F. (2013). Functional connectivity using BrainStorm, OHBM 19, Seattle, WA, USA. • Joshi, A. A., Ashrafulla, S., Shattuck, D. W., Damasio, H. & Leahy, R. M. (2013). Automated analysis of the shape of sulcal curves using the anisotropic Helmholtz equation. OHBM 19, Seattle, WA, USA. • Joshi, A. A., Ashrafulla, S., Shattuck, D. W., Damasio, H. & Leahy, R. M. (2012). An invariant shape representation using the anisotropic Helmholtz equation. MICCAI, 15(3), 607–14. This work was funded under the NIBIB T32EB00438 Bioinformatics Training Program and NIH grants R01EB009048 and R01 5R01EB000473. http://ashraful.lasyedashrafulla@gmail.com

More Related