Want to get involved? We're always looking for ideas and content for Weekly Challenges.
SUBMIT YOUR IDEAYuck on this challenge!
Whoa, total DS newbie here. I definitely need to learn more about using the predictive tools.
I played around with using different modeling tools... but I wouldn't have survived without everyone's solutions and online "lit review."
This was a cool practice - I don't understand the foundations of why we use certain models, but here's my train of thoughts:
I saw that both a Forest Model and Boosted Model produced Variable Importance Plots. However, the Boosted Model plot didn't have F_38 in its Top 10, so I went with a Forest Model.
With the Forest Model, I went with Logistic Regression because a Linear Regressions didn't allow me to select H0 as a target variable.
tldr: This was definitely a total newbie blind tool grasping/process of elimination challenge for me. 😛
A lot of googling for this challenge!
Definitely I have to come back to this one. Couldn't find anything in the community on how to estimate the chi-sq effect apart from the contingency table tool. Then I looked at the solution and found out about the Nested Test tool, but still I have some studying to do here