Not sure exactly what you mean, but if you're talking about the mean absolute deviation you use absolute value of the difference. So, in that sense, you only care about the difference between the value and the mean. Since it's thought of as a "distance" from the mean, that's why we use absolute value, which makes the distance always positive.
Karl Fisch ...... Can you please describe how you calculate the mean absolute deviation ? I don't understand exactly how you ignore the negative values... ?
Since we are calculating the distance from the mean, we only care about the distance, not whether it's "over" or "under". Therefore, we use the absolute value of the difference. Absolute Value always makes it positive, so that we treat both "over" and "under" the same, allowing us to calculate how spread out the data is.
