Problem 1: (50 pts) You are given injury severity data for 2,273 single-vehicle motorcycle crashes in Indiana. There are four possible severity outcomes: no injury (property damage only and possible injury), non-incapacitating injury, incapacitating injury, and fatality. You want to know what the likelihood of an individual being involved in a crash is based on the available crash data characteristics. Your task is to estimate an ordered probability model of motorcyclists' injury severity. • Following a forward stepwise process, find your best fit model specification. Some things to consider as you fit your model: Make sure the dependent variable is ordered and follows the format required for model estimation. - Do the signs of the coefficients make sense? - Is there acceptable correlation among explanatory variables? You must create some indicators that I did not cover in the tutorial documents. - Do you need to correct for heteroskedasticity? • After you have arrived at your best fit model specifications, provide a detailed discus- sion of the logical process that led you to the selection of your final specification. Present the descriptive statistics of the variables included in your final model specifica- tions as shown in the document on Canvas. You do not have to categorize the variables by category, but are welcome to do so. Are there any statistics worth highlighting? Points will be deducted for incorrect descriptive statistics presentation. • Present your model as shown in the document on Canvas. You do not have to categorize the variables by category, but are welcome to do so. Points will be deducted for incorrect model presentation. • Provide a discussion for each of the variables in your final model specifica- tion. Based on your model, what are the effects of the variables on fatal/injury crash probability? What are some plausible reasons for the significance of the variable and its effect on injury severity outcome? You are welcome to use your intuition, but also encouraged to find sources that confirm/validate your results. • Summarize your findings and provide some potential solutions for at least three of the variables in your final model specification. Use the countermeasure selection resources to determine your proposed solutions, explain why a countermeasure may be effective, and what the anticipated increase in safety would be should the countermeasure be implemented. • Prepare your deliverable as a mini-report based on what is being asked in the bullet points above. Definitions of available variables are given on the following page.

Solved: Problem 1: (50 pts) You are given injury severity data for 2,273 singl

Home
questions and answers
problem 1 50 pts you are given injury severity data for 2 273 single v

Question

Problem 1: (50 pts) You are given injury severity data for 2,273 single-vehicle motorcycle crashes in Indiana. There are four possible severity outcomes: no injury (property damage only and possible injury), non-incapacitating injury, incapacitating injury, and fatality. You want to know what the likelihood of an indiv