Fixed Activity Prism Generator

The following prism vertices are determined for all individuals, both adults and children using stochastic frontier models:

a. Earliest start time of day (start of an out-of-home activity)
b. Latest end time of day (must arrive home by this time)

A negative binomial regression model is then run to determine the number of work/school episodes for individuals who have jobs and children attending school. For each work/school episode the following prism are generated:

a. Latest arrival time at work/school
b. Earliest departure time from work/school

Separate models may need to be estimated for first and second jobs in case of workers with multiple jobs. There also need to be consistency checks put in place to ensure that the vertices of the prisms of different episodes do not overlap.

-- KarthikKonduri - 15 Feb 2010