You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
These two weren't ever defined. What do people think? @jbpoline, @gllmflndn?
nidm:effectDegreesOfFreedom: Degrees of freedom of the effect.
In the context of a general linear model, the effect degrees of freedom is the rank of the contrast. For example, a contrast comprising of a vector has effect degrees of freedom of 1 and can be tested with a F-test with numerator degrees of freedom of 1.
nidm:errorDegreesOfFreedom: Degrees of freedom of the error.
In the context of a general linear model, the error degrees of freedom are the number of observations less the rank of the design matrix.
Note the awkward description of "contrast consisting of ... vector"; we have not defined whether a contrast matrix consists of rows or columns of contrast vectors, and hence I left it vague.
The text was updated successfully, but these errors were encountered:
The suggested update was implemented in #318 where the term status was set to "pending final vetting". This issue remains open to continue the discussion until we reach a consensus.
These two weren't ever defined. What do people think? @jbpoline, @gllmflndn?
In the context of a general linear model, the effect degrees of freedom is the rank of the contrast. For example, a contrast comprising of a vector has effect degrees of freedom of 1 and can be tested with a F-test with numerator degrees of freedom of 1.
In the context of a general linear model, the error degrees of freedom are the number of observations less the rank of the design matrix.
Note the awkward description of "contrast consisting of ... vector"; we have not defined whether a contrast matrix consists of rows or columns of contrast vectors, and hence I left it vague.
The text was updated successfully, but these errors were encountered: