18
It follows that
Aj⅝' ≥ λicbii ⅛ λid - λ7c∣δn∙∣
= c(λiba — λj ∖bji I) + λid
> Xid
> 0.
Hence we have
On the other hand, it is not difficult to see that the procedure will terminate
within a finite number of iterations since each entry ‰7∙ is hnite. Hence the procedure
produces the matrix A = AU in standard form where U is a unimodular matrix.
We complete the proof. □
We remark that the origin of Rn is still contained in the interior of the convex
hull of the vectors ɑɪ,..., an+1. Moreover, the corresponding convex combination
coefficients remain unchanged. It is also easy to show that the volume of the simplex
does not change under unimodular transformation. For n = 2, we can construct
such unimodular matrices by adapting Scarf’s method in [10]. The procedure now
can be applied.
Let us give some examples.
Example 5. We are given
0 -1
1 1
-1 0
Then
1 1
-1 0
More intriguing information
1. Orientation discrimination in WS 22. Categorial Grammar and Discourse
3. Do the Largest Firms Grow the Fastest? The Case of U.S. Dairies
4. Analyzing the Agricultural Trade Impacts of the Canada-Chile Free Trade Agreement
5. The name is absent
6. Kharaj and land proprietary right in the sixteenth century: An example of law and economics
7. The name is absent
8. Density Estimation and Combination under Model Ambiguity
9. PRIORITIES IN THE CHANGING WORLD OF AGRICULTURE
10. The name is absent