A_A_B_testing | statistical analysis , a dataframe

A_A_B_testing is a Jupyter Notebook library. A_A_B_testing has no bugs, it has no vulnerabilities and it has low support. You can download it from GitHub.

During the statistical analysis, a dataframe was processed containing the activity of users of the mobile application for food delivery. Data processing was carried out, data verification was carried out, a funnel of events was built and the results of A/A/B testing were analyzed. Results: 1. Duplicates have been deleted, no gaps have been found in the data. 2. The date and time format is given in an acceptable form. 3. The time interval in the analyzed data: from August 1 to August 7 (taking into account the removal of unnecessary data). 4. The funnel of events looks like this: MainScreenAppear OffersScreenAppear CartScreenAppear PaymentScreenSuccessful. 5. The greatest loss of users occurs when switching from the main screen to the list of services 6. The share of users who switch from the main screen event to the payment of the goods is 48 percent. 7. The breakdown of the control groups was carried out correctly, the data were compiled correctly. 8. The number of tests performed when comparing each group is 5. The total number of comparisons is 20. Therefore, the significance level is 0.0025 (according to Bonferoni) 0.0026 (according to Shidak). 9. There is no statistically significant difference between the groups with the original font of the application design and the modified font of the design, therefore, you can stop the experiment and admit that there is no difference in the font of the design. 10. To confirm the conclusion, Bonferroni and Shidak were checked and the result did not change.