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
According to what we found in #133, we should modify the current implementation of \tau and \rho so that they return (d,d) matrices on d-variate copulas, containing bivariate kendall and spearman coefficients. This will match correctly what is outputed by Statsbase.corkendall and Statsbase.corspearman. The fitting of archimedeans needs to be re-written a bit to take this into account..
Then the API could be exported.
The text was updated successfully, but these errors were encountered:
According to what we found in #133, we should modify the current implementation of \tau and \rho so that they return (d,d) matrices on d-variate copulas, containing bivariate kendall and spearman coefficients. This will match correctly what is outputed by
Statsbase.corkendall
andStatsbase.corspearman
. The fitting of archimedeans needs to be re-written a bit to take this into account..Then the API could be exported.
The text was updated successfully, but these errors were encountered: