Assignment No 3- CVL 910- F23

.pdf

School

Toronto Metropolitan University *

*We aren’t endorsed by this school

Course

910

Subject

Civil Engineering

Date

Jan 9, 2024

Type

pdf

Pages

3

Uploaded by AgentNightingalePerson973

Faculty of Engineering, Architecture and Science Department of Civil Engineering CVL910 – Transport Planning Assignment #3 : Mode Choice Modelling Due date: Friday November 3 , 202 3 at 11:59 PM The purpose of this assignment is to estimate a mode choice model of mode preferences using a Logit model. The dataset consists of a sample of trips collected from the TTS11 travel survey in the Greater Toronto and Hamilton Area (GTHA). The data provided contains information on the selected mode of travel and a set of socio-economic attributes of each trip. The table below explains each column in the data Table 1: ‘ tts11_tp3.csv’ Column Name Data Type Description values row Serial Row index choice Mode choice values 1: Public Transit (bus or subway) 2: Bicycle 3: Car 4: GO Transit gender binary 1: Female, -1: Male, 0: Unknown/Not specified driver_lic binary 1: user has driving licence, -1: no driving licence 0: Unknown pass_ttc binary 1: user has TTC pass, -1: no TTC pass, 0: Unknown pass_go binary 1: user has GO pass, -1 no GO pass, 0: Unknown emp_ft binary 1: full-time worker, -1: part-time worker, 0: Unknown/unemployed student binary 1: full-time student, -1: part-time student, 0: Unknown/not a student free_park binary 1: user has free parking at work/school, -1: no free parking, 0: Unknown/not applicable trip_type binary 1: home-based trip, -1: non-home-based trip, 0: Unknown/other
trip_purp categorical Purpose of trip 1: Home, 2: School, 3: Work, 4: Leisure/shopping 5: Other, 0: Unknown trip_dest_pd categorical Trip destination zone 1: Toronto Downtown Core, 2: Rest of South Toronto, 3: York, 4: East York, 5: North York, 6: Etobicoke 7: Scarborough, 8: Rest of GTHA, 0: Unknown trip_km scale Trip journey distance (km) Using BIOGEME software, develop a multinomial logit model for mode choice given this dataset. A skeleton code is provided to guide you in starting your code. Your submission on D2L must include: 1. One written report (12-page max) in .doc/.docx or .pdf format which includes: a) Signed report cover page https://www.ryerson.ca/civil/forms-resources/ b) BIOGEME estimation results for your final model specification (beta values, standard error, statistical significance, final log-likelihood, rho-square value) [3 marks] c) A discussion on these results (800 words max) [3 marks] d) Compare the differences and advantages of using a discrete choice model over a linear regression model? [2 marks] e) Describe the process which your final model was obtained (variable selection, preliminary models, rejected parameters, etc.) [5 marks] f) Any additional variables that you would like to have included in the model that may improve it. [1 mark] [Total 14 marks] 2. One working script/code (with comments explaining your logic) [6 marks]
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help