AB Testing
-
Strategy
Python, Pandas, Scipy, Statsmodels
-
Design
Data Science, Statistical Analysis
-
Tags
Business Problem
A fast food chain is planning to add a new product to its menu.
They are undecided about three possible marketing campaigns to promote this new product.
To determine which promotion has the greatest impact on sales, the new product is promoted in several randomly selected convenience stores.
With these promotions in different locations, the weekly sales of the new product are recorded for the first four weeks.
Evaluate the results of the A/B test and decide which marketing strategy works best.
Dataset Info
Total Features: 7
Total Row: 548
Feature | Definiton |
---|---|
MarketID | Unique market id |
MarketSize | Market Size |
LocationID | Unique store location |
AgeOfStore | Age of Store |
Promotion | Type of promotion tested |
week | One of the four weeks when promotions are managed |
SalesInThousands | Sales Amount |
Requirements
matplotlib==3.5.2
numpy==1.21.5
pandas==1.4.3
scipy==1.7.3
statsmodels==0.13.2
Files
ab-testing.ipynb - AB Test Project with proportion Z test.
ab-testing-anova.ipynb - AB Test Project with ANOVA