In this repository you can find code for hypothesis testing on mtech data. To see the web application you can follow this link.

All underlying code and explanations you can check in this notebook. The project consists of several parts. In the beggining you can upload your csv file and test two of the hypothesis:

1. Men miss more than 2 working days (work_days) during the year due to illness significantly more often than women
2. Workers over 35 years of age (age) miss more than 2 working days (work_days) due to illness during the year significantly more often than their younger colleagues

All corresponding distribution graphs and results of testing you can see on web application. Also I implemented Bootstrapping to count the significant interval of missed days both for two groups.