Modeling correlation in a sequential link choice model
Project Number: 207
Project Leader: Emma Frejinger
Start Date: 0000-00-00
End Date: 0000-00-00
Description:Research Question What is the best way to model correlation in a sequential link choice model; A nesting approach with closed form probabilities or with link specific error components and hence non-closed form probabilities? How different are the prediction results and how do these compare to the route choice logit where the independence of irrelevant alternatives property holds?