-cmxtmixlogit- was newly introduced in Stata 16 to enable choice modelling for panel data. After watching Joerg Luedicke's excellent webinar on the topic & having worked through the 8 introduction chapters in the documentation, one question about -cmxtmixlogit-'s application still nags at me - namely, where does one access such granular data in real life examples?
To illustrate my point, consider the fictitious dataset used in the documentation:
Code:
. use http://www.stata-press.com/data/r16/transport.dta, clear (Transportation choice data) . list if id <= 2, sepby(t) +------------------------------------------------------------------------+ | id t alt choice trcost trtime age income parttime | |------------------------------------------------------------------------| 1. | 1 1 Car 1 4.14 0.13 3.0 3 Full-time | 2. | 1 1 Public 0 4.74 0.42 3.0 3 Full-time | 3. | 1 1 Bicycle 0 2.76 0.36 3.0 3 Full-time | 4. | 1 1 Walk 0 0.92 0.13 3.0 3 Full-time | |------------------------------------------------------------------------| 5. | 1 2 Car 1 8.00 0.14 3.2 5 Full-time | 6. | 1 2 Public 0 3.14 0.12 3.2 5 Full-time | 7. | 1 2 Bicycle 0 2.56 0.18 3.2 5 Full-time | 8. | 1 2 Walk 0 0.64 0.39 3.2 5 Full-time | |------------------------------------------------------------------------| 9. | 1 3 Car 1 1.76 0.18 3.4 5 Part-time | 10. | 1 3 Public 0 2.25 0.50 3.4 5 Part-time | 11. | 1 3 Bicycle 0 0.92 1.05 3.4 5 Part-time | 12. | 1 3 Walk 0 0.58 0.59 3.4 5 Part-time | |------------------------------------------------------------------------| 13. | 2 1 Car 0 4.36 0.23 3.0 2 Full-time | 14. | 2 1 Public 0 4.43 0.43 3.0 2 Full-time | 15. | 2 1 Bicycle 0 1.25 1.23 3.0 2 Full-time | 16. | 2 1 Walk 1 0.89 0.12 3.0 2 Full-time | |------------------------------------------------------------------------| 17. | 2 2 Car 0 7.14 0.23 3.2 3 Part-time | 18. | 2 2 Public 1 1.54 0.12 3.2 3 Part-time | 19. | 2 2 Bicycle 0 2.75 0.95 3.2 3 Part-time | 20. | 2 2 Walk 0 0.53 1.64 3.2 3 Part-time | |------------------------------------------------------------------------| 21. | 2 3 Car 0 6.69 0.17 3.4 2 Full-time | 22. | 2 3 Public 1 1.32 0.34 3.4 2 Full-time | 23. | 2 3 Bicycle 0 0.60 0.49 3.4 2 Full-time | 24. | 2 3 Walk 0 0.68 0.63 3.4 2 Full-time | +------------------------------------------------------------------------+
When I think about how this data might be constructed in real life, I suppose we can survey the same commuters over time, ask for their range of transportation options, age, etc. & then observe and record the mode actually chosen. However, how about data on the cost & time for options such as Bicycle or Walk that are never taken by a given individual? Do we ask the survey respondent to speculate on, say, how much productivity would be lost if he walked instead of drove to work in determining the value for -trcost-? And, if so, how much confidence can we have in alternative-specific values pertaining to alternatives that are never exercised?
Question
I would therefore like to ask if other researchers apply something akin to -cmxtmixlogit- in their work to help develop my intuition on this topic a bit more. What are some examples of real-life research questions that enable its application?
Thanks.
0 Response to cmxtmixlogit: real life data & applications?
Post a Comment