Detecting and controling outliers with PowerBI

[@msughmosesatser5306]

Thank you so much. I never seen a very simple way of controlling outliers in your dataset as good as this.

[@AnalyzeIt_Nael]

Glad it was helpful!

[@AnalyzeIt_Nael]

Also please note that z-score method assumes the data sets has a frequency distribution that are mound shaped and symmetric.

[@srinivasseena6574]

Superb

[@elahehsamandi914]

Thank you, It was very helpful, and you explained it excellently!

[@AnalyzeIt_Nael]

You are welcome

[@08152713755]

Thanks. This is what i have been looking for

[@AnalyzeIt_Nael]

Glad, it helps you in your job. You can also see the outliers using an imported box plot visual powerbi.microsoft.com/en-us/blog/visual-awesomeness-unlocked-box-and-whisker-plots/ Power bi also integrates Python and if you know package such as numpy, matplot lib, you can get a lot of these done visually too

[@TassouIffa]

Many thanks ! good explanation

[@stephaniewilson5595]

Excellent video , thank you!

[@hannaberndt8249]

Thank you - this is really super helpful!

[@chukkyochonogor7601]

Thanks, very helpful

[@kittydoes7600]

Is this the same process if I have months as slicers and want to filter outliers based on selected months?

[@Xiao-rf9kl]

Hello, how about if I want dynamic zscore based on data slicers? Thanks!

[@AnalyzeIt_Nael]

I would think that you can have the mean and standard deviation as calculated measures and then you do your z-score as a new column on which you can use a slicer

[@MAbdullah47]

where we can find the resources of this Video?

[@priyankajadhav6742]

"no ole db error information found " i am getting the above error . can you help?

[@tameimpalan]

why did u used -3 and 3? could u explain that pls?

[@AnalyzeIt_Nael]

Let me first emphasize that there are serval ways in which you can determine if an observation is an outlier. z-score method only applies to data sets with frequency distributions that are mound shaped and symmetric. The z-score tells you how many standard deviations away a value is from the mean. Any z-score greater than 3 or less than -3 is considered to be an outlier. This rule of thumb is based on the empirical rule. From this rule, it is often seen that almost all of the data (99.7%) would be within three standard deviations from the mean. By calculating the z-score you are standardizing the observation, meaning the standard deviation is now 1. Thus, from the empirical rule you expect that 99.7% of the z-scores to be within -3 and 3. Another method you can use to determine the Interquartile Range (IQR) outlier Fences.

[@tameimpalan]

@@AnalyzeIt_Nael Amazing explanation, thank you so much!