PREDICTIONS OF INTERCITY MODAL CHOICE FROM DISAGGREGATE, BEHAVIORAL, STOCHASTIC MODELS

Abstract

THIS PAPER REPORTS THE RESULTS OF SOME EMPIRICAL PREDICTION TESTS ON DISAGGREGATE, BEHAVIORAL, STOCHASTIC MODELS OF MODAL CHOICE. IT IS DISCOVERED THAT THE EXPECTED-NUMBER CALCULATION (SUMMATION OF PROBABILITIES) YIELDS MODAL-CHOICE PREDICTIONS WITH A HIGH DEGREE OF ACCURACY, ALTHOUGH A CAVEAT IS IN ORDER REGARDING THE TRANSFERABILITY OF SUCH MODELS: THEY SHOULD BE TRANSFERRED WITH CARE AND ONLY TO SIMILAR SITUATIONS. THE STABILITY OF THE MODELS IS EVIDENCED BY THE FACT THAT THEY CAN BE BROKEN DOWN BY TRIP PURPOSE AND REAGGREGATED WITHOUT LOSS OF PREDICTIVE POWER. FINALLY, THE METHOD OF OBTAINING PREDICTIONS BY CLASSIFYING PROBABILITIES IS CONSIDERED AND REJECTED.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.