THIS PAPER GIVES THE APPROXIMATE JOINT PROBABILITY DISTRIBUTION FUNCTION OF THE LINK TRAFFIC FLOWS ON A NETWORK. THE KNOWLEDGE OF SUCH DISTRIBUTION FUNCTION ENABLES US TO DERIVE THE LOG-LIKELIHOOD FUNCTION FOR THE ESTIMATION (WITH LINK DATA) OF THE TRAFFIC DIVERSION PARAMETER, THETA, THAT APPEARS IN MOST STOCHASTIC ASSIGNMENT MODELS. IT ALSO ENABLES US TO PERFORM GOODNESS TO FIT STATISTICAL TESTS FOR THE VALIDATION OF TRAFFIC ASSIGNMENT METHODS. BOTH OF THESE PROCEDURES ARE ILLUSTRATED WITH A SMALL NUMERICAL EXAMPLE. IT IS ALSO POINTED OUT THAT IT IS POSSIBLE TO CALIBRATE AND/OR VALIDATE A TRAFFIC ASSIGNMENT MODEL WITH INCOMPLETE DATA BASES (I.E. WHEN TRAFFIC COUNTS ARE ONLY AVAILABLE ON A FEW OF THE LINKS OF THE NETWORK).(Author/publisher).
Abstract