Dear List,
I am currently using a mixed logit to estimate wtp to avoid power outages. The attributes are duration of power outage, time of day of outage (peak versus non-peak), season in which outage occurs and day (weekday versus weekend).
I have tried using a log normal distribution, but as noted by several authors on this subject my WTP estimates are huge, particularly for duration ~£195.
I would like to re-estimate the model specifying a triangular distribution for duration.
How can I incorporate a triangular distribution in mixlogit? It is easy to incorporate the log normal, but I cannot find the equivalent for the triangular distribution.
Any help would be great.
Kind regards,
Karyn
I am currently using a mixed logit to estimate wtp to avoid power outages. The attributes are duration of power outage, time of day of outage (peak versus non-peak), season in which outage occurs and day (weekday versus weekend).
I have tried using a log normal distribution, but as noted by several authors on this subject my WTP estimates are huge, particularly for duration ~£195.
I would like to re-estimate the model specifying a triangular distribution for duration.
How can I incorporate a triangular distribution in mixlogit? It is easy to incorporate the log normal, but I cannot find the equivalent for the triangular distribution.
Any help would be great.
Kind regards,
Karyn