In AHP method of weighting, as the number of variables increases so should the Consistency ratio. Prof Saaty suggested a CR of 0.1 irrespective of the number of variables, researchers have found this to be too restrictive. the formulation by LAMATA gives a formulation to adapt to the number of variables but how many variables for what CR?
you put your fingers at a sensible part of the MCE, particularly AHP.
The 'traditional' criterion of accepting matrices is sometimes critizesed due to
their inflexibility and because it may be too restrictive when the size of the matrix increases.
Alternative approaches are e.g. discussed by Aguarón J. and Moreno-Jiménez J M (2003). European Journal of Operational Research, 16 May 2003, Volume: 147 Issue: 1 pp.137-145.
out of the many examples one may also mention Ergu et al. (2011) hwo proposed a maximum eigenvalue threshold as the new consistency index for the data in the AHP and ANP, which is mathematically equivalent to the CR method.
Alonso& Lamata (2006) developed a system compares the matrix level of consistency to the consistency level of the remaining matrices (random matrices) of the same dimension. see: http://hera.ugr.es/doi/16515833.pdf
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size