I have many variables as the causes of traffic congestion, which I subjected to ranking by their level of significance and contribution to traffic congestion. Which of the ranking analysis is best fit for this situation?
Thank you.
Do you use an econometric model ? If so, the estimated weights of each variable (and their respective T stats) could be helpful
I recommend also what Mr. Vsevolod said. In addition, Pearson correlation between the concerned variables and the traffic congestion can also be used (This can be done using any statistical software)
You can also try to pick up the regression equation and the value of the coefficients of the equation to make conclusions about the importance of each factor.
Since you seem to have a binary outcome (congestion yes or no), I would do a logistic Regression and take the sequence of entry of the variables. But watch out for high correlations between the predictors.
A few days before, I have done a work related to variable selection. I used weka data mining software to do that. In weka there is seven evaluation methods and a ranker algorithm. You can use the combinations of evaluation method and ranker. Each combination will give you list of the variables according to their importance.
I think the following link can be a help for you:
http://www.cs.waikato.ac.nz/ml/weka/mooc/moredataminingwithweka/slides/Class4-MoreDataMiningWithWeka-2014.pdf
- Badges
- Science topic
- Similar topics
- Engineering
- Civil Engineering
- Transportation Engineering