ConstrainedMin and vector-notation
- To: mathgroup at smc.vnet.net
- Subject: [mg21870] ConstrainedMin and vector-notation
- From: nielses at my-deja.com
- Date: Wed, 2 Feb 2000 22:54:26 -0500 (EST)
- Sender: owner-wri-mathgroup at wolfram.com
I have the following problem: (used to calculate the Farell efficiency
measure, for those interested)
min l
s.t.
t(X) z <= l y
zi >= 0
z1 + ... + zn = 1
X: matrix, z, y: vectors, l: scalar
(t(X) is the matrix X transposed)
I want to solve this with ConstrainedMin. (I realize it can be rewritten on
matrixform and fed to LinearProgramming.)
The following works fine, but I wonder, if it's not possible to write it in an
easier way:
X = {{1, 3}, {7, 1}}
y = {8, 4}
n = Length[X];
zz = Table[z[i], {i, n}];
ConstrainedMin[l,
Join[MapThread[LessEqual, {zz.X, l y}],
Table[z[i] >= 0, {i, n}], {Plus @@ zz == 1}], Append[zz, l]]
The point is of course, that is has to work with
X = {{3, 6}, {4, 4}, {8, 2}}
too, and an y-vector of a higher dimension for that matter.
Can anyone help me write this more elegantly??
I often encounter situation where I have to define a vector like zz to obtain
a vector-solution (e.g. when using Solve). There has to be a simpler way.
Regards,
Niels Elken Sønderby
nielsx at bigfoot.com
Sent via Deja.com http://www.deja.com/
Before you buy.