So what can you change to improve a Model ‘s fit to data?
Different deterrence functions
- negative power
- negative exponential
Different sets of Constraints
- unconstrained
- origin constrained
- destination constrained
Different Methods of Calibration
- change Beta to minimise sum of errors squared
- change Beta to minimise sum of absolute errors
- change Beta to obtain the same mean trip cost as the observed data
(called Entropy Maximising in AGW)