Variance Estimation with Mean Absolute Deviation (MAD)

Non-Causal Step

This step applies the Mean Absolute Deviation (MAD) algorithm to estimate the variance of the data. The basic formula is as follows:

$${\hat{\sigma} = K \cdot \text{MAD}}\quad\text{,}$$

where $K \approx 1.4826$, MAD is a measure returned by the Mean Absolute Deviation (MAD) algorithm and $\hat{\sigma}$ denotes the estimated variance.

Input Parameters

1. Input data

Output Parameters

1. Estimated variance

Workflow

Algorithm

Mean Absolute Deviation

References

• D.C. Hoaglin, F. Mosteller, J. W. Tukey, Understanding Robust and Exploratory Data Analysis, John Wiley and Sons, 1983.