Suppose I have a vector of positive weights a=(a1, a2, a3, a4) such that a2=a3 and a1+a2+a3+a4=1. Is there any way to sample this kind of weights using R? I tried to think about using the Dirichlet distribution, but it gives no mechanism to force two of the variates to be equal.
Aucun commentaire:
Enregistrer un commentaire