search for: ipardo

...each model can be correlated. My understanding is that although lm() handles vector Y's on the left-hand side of the model formula, it really just fits m separate lm models. What should I use to do a full multivariate analysis (as in section 7.7 of Johnson/Wichern)? Thanks. Iain Pardoe <ipardoe at lcbmail.uoregon.edu> University of Oregon

...-m)) * F.quant But, what if instead I'd like to sample (y1,y2) values from this distribution? I can sample from an F(m,n-r-m) distribution easily enough, but then how can I transform this to a single point in (y1,y2) space? Any ideas would be gratefully appreciated. Thanks. Iain Pardoe <ipardoe at lcbmail.uoregon.edu> University of Oregon