Median Absolute Deviation (Around the Median)

The Median Absolute Deviation Around the Median is a measure of dispersion of a set of data points. It can be used in place of the standard deviation when the weight of outliers shall be reduced.

Indicator Description

The Median Absolute Deviation Around the Median is a measure of dispersion of a set of data points. It can be used in place of the standard deviation when the weight of outliers shall be reduced.

Calculation: The Median Absolute Deviation of the selected data set is the median of the absolute deviations from the median of the data set.

The Median Absolute Deviation is calculated from price data over the selected lookback period.

  1. We start with the calculation of a central tendency, such as the mean, median or mode of the selected input data.
  2. Next we find out how much the data points deviate from the central tendency and calculate the absolute differences dropping any negative signs.
  3. We take the absolute values of the differences and calculate the median.

The indicator which can be downloaded here is the median absolute deviation around the median, which is often referred to as the “median absolute deviation”.

This calculation is a much more robust measure of dispersion than the standard deviation. Large outliers will have a far lesser impact on the dispersion, when the mean absolute deviation around the median is used instead of the standard deviation.

The Median Absolute Deviation around the median is always smaller than the standard deviation. When the data is normally distributed the standard deviation is about 1.4826 times the Median Absolute Deviation. The exact factor can be derived from the cumulative distribution function of the standard normal distribution.

The indicator is available for NinjaTrader 8.