How can I statistically measure/determine if A performs better than B?
Hi Data Science Community!
I am a new Data Intern and I have been stuck on this question for a while.
Here is a sample dataset I am working with:
| Customer | Manufacturer A Spending | Manufacturer B Spending | Manufacturer A Cost per Product (CPP) | Manufacturer B Cost per Product (CPP) | Product Cost Difference (B-A) | Product Cost Difference in % |
|---|---|---|---|---|---|---|
| 1 | 400000 | 360000 | 44 | 45 | 1 | 1/45 |
| 2 | 300000 | 310000 | 23 | 21 | -2 | -2/21 |
| 3 | 100000 | 106000 | 1.4 | 1.6 | 0.2 | 0.2/1.6 |
- I have 35 customers, ordered by Spending DESC.
- Spending for customers ranges from as high as 400000 to as low as 2000.
- You can assume that Manufacturers A and B produce the same goods for the same customer. (Different for each Customer)
- Spending for each Manufacturer are not the same but very close to each other.
- You can assume each Manufacturer produce only 1 Product for each customer.
I would like to test if Manufacturer A is better than Manufacturer B.
I added the additional column of the Cost Difference in % as it gives a better representation of the difference in Cost per Product since each customer's cost is different. - For the first customer, Manufacturer A is 1/45 better(cheaper) as compared to Manufacturer B.
Intuitively I could simply plot a histogram or bell curve for visualization and perform a 1 tailed Z-Test to determine if A is better than B. However, I believe that I have to consider each customer's spending as well since I have a huge difference between the largest spending customer and lowest spending customer.
I am thinking of considering a weighted Product Cost Difference but I am not sure if I am heading towards the right direction and how to do that.
I would appreciate any help or advice!!
Topic hypothesis-testing data
Category Data Science