Hampel Filter and constant data

1. How does a Hampel filter work?
2. How do you use Hampel?
3. How do you filter outliers in a time series?
4. Which outlier detection should I use?

How does a Hampel filter work?

The Hampel Filter block detects and removes the outliers of the input signal by using the Hampel identifier. The Hampel identifier is a variation of the three-sigma rule of statistics, which is robust against outliers.

How do you use Hampel?

Use hampel to locate every sample that differs by more than three standard deviations from the local median. The measurement window is composed of the sample and its six surrounding samples, three per side. [y,i,xmedian,xsigma] = hampel(x); Plot the filtered signal and annotate the outliers.

How do you filter outliers in a time series?

A simple method to detect outliers is to estimate the rolling center of the time series by fitting a smooth curve to the series. You can then classify an observation as an outlier if it is sufficiently far away from the curve.

Which outlier detection should I use?

The two main types of outlier detection methods are: Using distance and density of data points for outlier detection. Building a model to predict data point distribution and highlighting outliers which don't meet a user-defined threshold.