Robust statistics are a type of statistics that are good at dealing with "outliers" in data. Outliers are values that are much larger or smaller than the other values in the data.
Imagine you have a set of numbers like this: 1, 2, 3, 4, 5, 1000. The numbers 1, 2, 3, 4, and 5 are all close to each other, but the number 1000 is much larger than the others. This is an outlier, and it can have a big effect on the average (or mean) of the numbers. For example, the mean of this set is 201, because you get this by adding up all the numbers and dividing by 6. But if you remove the outlier (the number 1000), the mean becomes 3, because you get this by adding up the numbers 1, 2, 3, 4, and 5 and dividing by 5. So the outlier has a big effect on the mean.
Robust statistics are good at dealing with outliers because they are not very sensitive to the effects of a few extreme values. This means that they can still give you a good estimate of the "center" of the data, even if there are some outliers. So if you have a set of numbers with some outliers, using robust statistics can give you a better idea of what the data is really like.
So in cases like these we better use the median value to make results.
No comments:
Post a Comment