Skip to content

The Data Scientist

the data scientist logo

Performance measures: standard deviation, MAD, and quantiles


Wanna become a data scientist within 3 months, and get a guaranteed job? Then you need to check this out !

xSomething else which I have found helps a lot is the use of error histograms and the standard deviation or quantiles when reporting errors. So, for example,

<GRAPH>

When a graph is relatively normal, the use of standard deviation makes sense, otherwise you might want to go for quantiles, or the median absolute deviation (MAD).


Wanna become a data scientist within 3 months, and get a guaranteed job? Then you need to check this out !