On some possible extensions of the Brodsky-Lepage-Mackenzie approach beyond the next-to-leading order

Abstract

Noting that the choice of renormalization point advocated by Brodsky, Lepage and Mackenzie (BLM) is the flavor independent prescription which removes all f-dependence from the next-to-leading order coefficients, we consider the possible generalization which requires all higher order coefficients ri to be f-independent constants ri*. We point out that in QCD, setting ri = ri* is always possible, but leaves us with an ambiguous prescription. We consider an alternative possibility within the framework of the BLM approach and apply the corresponding prescription to the next-to-next-to-leading approximation of [sigma]tot(e+e--->hadrons) in QCD. The analogous questions and the special features of the BLM and effective charge approaches in QED are also discussed.