Regression Analysis is one of the foundational concepts of predictive analytics. However, it is also one of the most underrated algorithms out there.
This talk is all about the forgotten beauty of regression analysis and how to leverage it for doing better data science.
This talk goes into details about where regression analysis should be used, where it shouldn't be used, and some more important and beautiful topics in regression analysis which most of the popular courses in data science do not cover.
This talk is divided into the following parts:
Understanding Regression Analysis: This gives a quick primer about what regression analysis is, for the uninitiated audience.
The good: This part of the talk goes into details of the positives of regression analysis and the scenarios where one should/can leverage the power of the algorithm
The bad: This part of the talk goes into details of the positives of regression analysis and the scenarios where one should/can not leverage the power of the algorithm. and the better alternatives which exist in the data science world.
The untold: This part of the talk is about the important topics inside regression analysis which are not covered by any of the major data science courses available. This includes distribution analysis, residual analysis, and understanding non-linear dependencies and trends from the regression curve