Author: Guilherme Alves Ceobaniuk Zaluchi <guilherme@zaluchi.com.br>
Means are a very usefull statistic resource to measure a tendence, inclination or the variation of some function. However, means are known by having no precision at data representation. That's because it's a representation of the whole, with its variations, as if it were only one element repeated several times.
The proposed solution is the use of relative means. Relative Means are means relative to the own mean of the set.
- The low mean is the mean of the set of numbers that are lower than the mean of the whole set.
- The high mean is the mean of the set of numbers that are higher than the mean of the whole set.eto.
- The central mean is the mean of the set of numbers that are lower than the high mean, and higher than the low mean.
With these separations, and recursively re-applying the definitions, it is possible to reach more aproximated results for representations of big datasets of high variety.
Follow example of code (in the folde código) for the relative means in the following languages:
- JavaScript (CommonJS)