I’d like to discover which “calendar factors” (e.g. day of the week, month) have the strongest relationship with whether or not a particular product is sold at least once. For days when the product was sold, I have data like this:

Date        Day of Week  Day Type  Month
01-01-2014  Wednesday    Weekday   January
01-13-2014  Monday       Weekday   January
02-09-2014  Sunday       Weekend   February
02-15-2014  Saturday     Weekend   February
02-16-2014  Sunday       Weekend   February
02-23-2014  Sunday       Weekend   February

If we pretend that this is the entire dataset, we see that the product was sold mostly on Sundays in February.

I’ve come up with a simple way to rank each individual calendar factor (e.g. month) as well as a few combinations (e.g. day type + month) by creating a ratio of “sale days” vs. the entire population: using the example data, Wednesday + January would have a rank of 1 / 5. However, I’m struggling to determine which factors and/or combination of factors are most significant without simply picking the highest-ranking one.

I feel like there must be a stronger mathematical approach to solving this kind of problem. Any ideas? Also, how would you classify this problem?


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.