Regarding control group seminar
4 views
Skip to first unread message
Arun Srikanth
Sep 16, 2009, 1:02:05 AM9/16/09
to con...@che.iitm.ac.in, controlg...@googlegroups.com, nkai...@iitm.ac.in, aru...@iitm.ac.in, na...@iitm.ac.in, sridh...@iitm.ac.in
Hearty welcome to control group members. The control group seminar is scheduled today at MSB 132 at 4 pm. The speaker is Gigi sebastian (Phd student under Arun Tangirala). The title and the abstract is given below . All the members are requested to attend.
Title: Quantitative analysis of directional strengths in multivariate proceses
Abstract: Identification and analysis of directed influences in multivariate systems is an important problem in many scientific areas. In recent works on neural structure determination, two measures, namely, partial directed coherence (PDC) and directed transfer function (DTF) that provide direct and total coupling strengths between a pair of signals, were introduced. These measures are based on joint stationary representations of a multivariate process. The DTF is known to give a normalized directed total influence of one variable on another, while the PDC qualitatively detect direct influences.
In my talk, a quantification of transfer of variations in one variable to another in a multivariate process based on a of VAR / VMA (vector autoregressive / vector moving average) model signal flow graph reduction of the multivariate process, will be presented along with the description of the above mentioned methods. Valuable insights into the structure of the process and the strengths of the connectivities are obtainable by the proposed methods.
Title: Quantitative analysis of directional strengths in multivariate proceses
Abstract: Identification and analysis of directed influences in multivariate systems is an important problem in many scientific areas. In recent works on neural structure determination, two measures, namely, partial directed coherence (PDC) and directed transfer function (DTF) that provide direct and total coupling strengths between a pair of signals, were introduced. These measures are based on joint stationary representations of a multivariate process. The DTF is known to give a normalized directed total influence of one variable on another, while the PDC qualitatively detect direct influences.
In my talk, a quantification of transfer of variations in one variable to another in a multivariate process based on a of VAR / VMA (vector autoregressive / vector moving average) model signal flow graph reduction of the multivariate process, will be presented along with the description of the above mentioned methods. Valuable insights into the structure of the process and the strengths of the connectivities are obtainable by the proposed methods.
Arun Srikanth
Sep 16, 2009, 1:07:56 AM9/16/09
to controlg...@googlegroups.com, con...@che.iitm.ac.in, na...@iitm.ac.in, sridh...@iitm.ac.in, aru...@iitm.ac.in, nkai...@iitm.ac.in
Dear All,
Hearty welcome to control group members. The control group seminar is scheduled today at MSB 132 at 4 pm. The speaker is Gigi sebastian (Phd student under Dr.Arun Tangirala). The title and the abstract is given below . All the members are requested to attend.
Title: Quantitative analysis of directional strengths in multivariate proceses
Abstract: Identification and analysis of directed influences in multivariate systems is an important problem in many scientific areas. In recent works on neural structure determination, two measures, namely, partial directed coherence (PDC) and directed transfer function (DTF) that provide direct and total coupling strengths between a pair of signals, were introduced. These measures are based on joint stationary representations of a multivariate process. The DTF is known to give a normalized directed total influence of one variable on another, while the PDC qualitatively detect direct influences.
In my talk, a quantification of transfer of variations in one variable to another in a multivariate process based on a of VAR / VMA (vector autoregressive / vector moving average) model signal flow graph reduction of the multivariate process, will be presented along with the description of the above mentioned methods. Valuable insights into the structure of the process and the strengths of the connectivities are obtainable by the proposed methods.
Regards,
S.Arun Srikanth.
Hearty welcome to control group members. The control group seminar is scheduled today at MSB 132 at 4 pm. The speaker is Gigi sebastian (Phd student under Dr.Arun Tangirala). The title and the abstract is given below . All the members are requested to attend.
Title: Quantitative analysis of directional strengths in multivariate proceses
Abstract: Identification and analysis of directed influences in multivariate systems is an important problem in many scientific areas. In recent works on neural structure determination, two measures, namely, partial directed coherence (PDC) and directed transfer function (DTF) that provide direct and total coupling strengths between a pair of signals, were introduced. These measures are based on joint stationary representations of a multivariate process. The DTF is known to give a normalized directed total influence of one variable on another, while the PDC qualitatively detect direct influences.
In my talk, a quantification of transfer of variations in one variable to another in a multivariate process based on a of VAR / VMA (vector autoregressive / vector moving average) model signal flow graph reduction of the multivariate process, will be presented along with the description of the above mentioned methods. Valuable insights into the structure of the process and the strengths of the connectivities are obtainable by the proposed methods.
S.Arun Srikanth.
Reply all
Reply to author
Forward
0 new messages