72,030 views
Course "Data Analyst": https://bit.ly/3UeT8yj We continue the series of webinars on statistics! We all know that statistics is such a multifaceted thing that the same hypothesis can be tested in completely different ways, and vice versa - use one method to analyze data in different tasks. In the last webinar, we discussed how not to get confused in regression and correlation tasks. This time we will talk about how to: ● compare several groups using a method such as t-test; ● survive if there are a lot of zeros in the data and the point above does not work (spoiler: using the Mann-Whitney U-test); ● improve prediction accuracy with Bootstrap. Code and data at the link https://yadi.sk/d/-TLefuXoV7Z7FQ?w=1 Learn Data Science with us: https://karpov.courses/