How can we check to see how much the presence of an outlier (or perhaps a few outliers) affect the values of the statistics (such as mean, median)?
How can we check to see how much the presence of an outlier (or perhaps a few outliers) affect the values of the statistics (such as
Most used statistics are Mean, Median and Mode.
Let us now discuss the effect of outliers on different statistic.
Mean: Mean is the average of all observations in the data set. Therefore, if there is a outlier in the dataset then mean is hugely affected as it is a measurement which include all the observations.
Take two datasets one without outlier another with outlier. One observation in the dataset is replaced by an outlier.
Without outlier : {1,2,2,2,3}
With outlier : {1,2,2,2,93}
The mean of the dataset without outlier = 2
The mean of the dataset with outlier = 20
So, we can see a huge difference between the mean without an outlier and with an outlier.
Step by step
Solved in 3 steps
