Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
tutorial:multivariateanalysis [2017/08/17 12:33]
127.0.0.1 external edit
tutorial:multivariateanalysis [2018/06/12 21:15] (current)
129.125.23.204
Line 2: Line 2:
 ======Multivariate analysis of electrophysiological data====== ======Multivariate analysis of electrophysiological data======
 =====Introduction===== =====Introduction=====
 +
  
 The objective of this tutorial is to give an introduction to multivariate analysis of electrophysiological data. Multivariate methods aim to find task-related features in the data which allows prediction of to which task single trials belong. Note that this is very different from classical statistical testing, where such features are identified by pooling over multiple trials and/or subjects and where features are typically assumed to be independent. In this tutorial, you will learn to apply standard classification algorithms such as the support vector machine to electrophysiological data. Furthermore,​ you will learn about the importance of regularization. Note that FieldTrip uses the external Donders Machine Learning Toolbox ([[https://​github.com/​distrep/​DMLT|DMLT]]) for its multivariate analyses. This toolbox requires at least MATLAB distribution 7.6.0.324 (R2008a). The objective of this tutorial is to give an introduction to multivariate analysis of electrophysiological data. Multivariate methods aim to find task-related features in the data which allows prediction of to which task single trials belong. Note that this is very different from classical statistical testing, where such features are identified by pooling over multiple trials and/or subjects and where features are typically assumed to be independent. In this tutorial, you will learn to apply standard classification algorithms such as the support vector machine to electrophysiological data. Furthermore,​ you will learn about the importance of regularization. Note that FieldTrip uses the external Donders Machine Learning Toolbox ([[https://​github.com/​distrep/​DMLT|DMLT]]) for its multivariate analyses. This toolbox requires at least MATLAB distribution 7.6.0.324 (R2008a).