In statistics, median is generally a measurement that is not impacted by outliers. Rarely - if ever - do we need to consider outliers and their possible impact on median. For an example, suppose that we do pullups daily and our median pullup number in each set is 9. Our best day we've ever had in a set is 21 pullups. While the 21 sounds significantly higher than the median value of 9, that number won't have an impact on median because it happened once. In the very exceptional cases that we do, we can get the median absolute deviation and filter from there. In the video, TSQL: How To Get the Median Absolute Deviation, we look at an example of doing this.
Some questions that are answered in the video:
For mastering data transformation from one form to another form, check out the highest-rated Automating ETL course on Udemy. For a coupon to the course, check out the trailer video on the channel SQL In Six Minutes.
Other languages make this calculation much easier. Python, R and Oracle all have this measure built in, so we'll rarely need to code this relative to the language that we're using. It's still helpful for us to know how to calculate this. This year, I used this calculation to do some analysis on CoVid data sets when looking at risk factors, CoVid and death rates. Before doing this analysis, I still had to convert some of the data accordingly. In general, we'll rarely use this because outliers seldom impact median - but we may want to look at a data set without those extreme values.
Are you looking for tech consultants that can assist with design and development? From building custom applications to working with existing infrastructure that's causing you trouble, we can get you connected to consultants who can assist. You can contact for assistance.
SQL In Six Minutes (YouTube) | SQL In Six Minutes (Odysee) | Automating ETL | T-SQL In 2 Hours | Consumer Guide To Digital Security
