I have never worked with a Binary dataset (1 and 0) (True or False) so I'm unsure what kind of statistical tests I should run to draw up simple conclusions.

I'm doing a data science project and the company wants to know/understand what features/characteristics are important to decide whether or not a customer will renew their lease (last column in Graph). They want me to just perform basic/simple probability and statistical analysis on the dataset.

I'm working in a Juypter notebook (pandas, seaborm, numpy, matplotlib, sklearn). I would appreciate it if someone could help lead me in a direction to what kind of simple analysis I can run on binary data.

Heres a sample of the table:

ID Age 20-29 Age 30-39 Age 40-49 Age 50 > Lease Length < 1 year Lease Length 1-2 Years Lease Length > 3 Years Late Payment No Fine Violations Credit Score Below 600 Renews
312 1 0 0 0 0 1 0 1 0 0 1
313 0 0 1 0 0 0 1 0 1 1 0
314 0 1 0 0 0 1 0 0 0 0 0
315 0 0 0 1 1 0 0 0 0 1 1
  • 1
    $\begingroup$ Not sure about basic/simple probability and statistical analysis in terms of understanding what features drive the outcome, but what you can do is to fit simple model (something like DecisionTrees) and then do feature explanation, either using native feature importance or something like shap which nice visuals and a lot of explanation capabilities. Not sure in terms of purely binary data, but I would definitely give it a try. link - github.com/slundberg/shap $\endgroup$ – Yaroslaw Homenko Mar 29 at 12:06
  • $\begingroup$ you can try correlation (or even categorical correlation) between random variables (so one can see how much a certain feature is correlated with desired outcome). For example simple inspection shows that younger age groups renew the plan (highly correlated to desired outcome) $\endgroup$ – Nikos M. Mar 29 at 16:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.