Embedded Files
Tags: A/B Test, Hypothesis Testing, Statistical Analysis, Data Preparation, Plotting
NumPy
Matplotlib
Pandas
Python
Jupyter Notebook
● GOAL
● GOAL
Prioritize hypotheses aimed at boosting revenue. Conduct an A/B test and analyze the outcomes to determine an optimal strategy for revenue growth.
● OVERVIEW
● OVERVIEW
Big Store is a leading online retail platform, constantly exploring strategies to increase revenue. After data preprocessing and preparation, the project is divided into two main parts:
Part 1: Prioritization of nine hypotheses using the ICE and RICE methods to assess their potential impact on revenue.
Part 2: Calculation and visualization of key metrics, followed by A/B testing to draw conclusions on the best course of action.
This analysis includes data on transactions and visits during the test period. The second part also involves a statistical analysis to determine the significance of differences in conversion rates and average order size, helping to establish whether the observed differences are statistically significant or due to random variation.
● HYPOTHESES TESTED
● HYPOTHESES TESTED
Hypothesis 1
Hypothesis 1
The conversion rates for both groups are equal, implying that any observed differences occur due to chance.
H₀: P₁ = P₂ - there is no difference in conversion rates between the two groups.
H₁: P₁ ≠ P₂ - there is a difference in conversion rates between the two groups.
Hypothesis 2
Hypothesis 2
The difference in average order size between the test groups is equal.
H₀: µ₁ = µ₂ - there is no difference in the average order size between the two groups.
H₁: µ₁ ≠ µ₂ - there is a difference in the average order size between the two groups.
To see the code, please visit my GitHub:
Category: Data Cleaning
Seaborn
Pandas
Python
Jupyter Notebook
Page updated
Google Sites
Report abuse