Lattice-subspaces and positive bases in function spaces (Q1417882)

Let \({\mathbb{R}}^{\Omega}\) be the vector lattice of all real-valued functions defined on a nonempty set \(\Omega.\) The following problem is discussed: Given a vector space \(X\) generated by a family \(x_{1},\ldots,x_{n}\) of linearly independent positive elements of \({\mathbb{R}}^{\Omega},\) does a finite-dimensional minimal subspace of \({\mathbb{R}}^{\Omega}\) exist which has a positive basis and contains \(X\)? It is proved that the answer is positive if and only if the convex hull \(K\) of the range of \(\beta(t)=\left( x_{1}(t),\ldots,x_{n}(t)\right) /\left( x_{1}(t)+\ldots+x_{n}(t)\right) \) is a polytope. Also, it is proved that \(X\) is a lattice-subspace (equivalently, \(X\) has a positive basis) if and only if \(K\) is an \((n-1)\)-simplex. In both cases, a concrete positive basis is constructed by using the vertices of \(K.\) Special attention is paid to the case where \(\Omega\) is a convex set and \(x_{1},\ldots,x_{n}\) are linear functions. This includes the case where the \(x_{k}\)'s are positive elements of a Banach lattice.