Can multiparticle correlations be described by present analytical QCD calculations?

Previous experimental studies of various correlation functions in angular intervals and their comparison with QCD calculations are summarized. A good description is only obtained for a globally normalized function (r). The hypothesis of local parton hadron duality (LPHD) seems to be well fulfilled in this case. For all differentially normalized functions, however, several disagreements are observed and the validity of LPHD is in question. It is argued (using Monte Carlo calculations) that LPHD could still be valid also in this case - after improving the perturbative QCD calculations. Only after the inclusion of full energy-momentum conservation in the calculations one will have a better handle to estimate, how far non- perturbative effects are spoiling the agreement with the data. (14 refs)