Despike Data

Despiking is useful for data which has many points with the same data value. Supervisor's despike feature applies small changes to the original data value to make them slightly different, without altering the data in a statistically significant way. For example, if you have three data points with the value 1.0, two of the values may be changed to 0.9999999 and 1.0000001, and the other sample is left as 1.0. Local means are used to determine the order of the data points to be altered.

Activity Steps

  1. Click the Data Loader Component in the Project Tree.
  2. Select the Despike tab at the bottom of the Main Canvas.
  3. Fill in the following columns in the Despike table for as many rows as required.
    • Enable
    • Assay
    • Domain – If left blank, all samples for the selected assay are despiked.
    • Despiked Column – Data Column Component to be despiked.
    • Max Distance – The maximum distance from each sample used to calculate the local mean used to order samples with the same value in the dataset. Default: 10.
    • Comment – Optional comment
  4. Click Despike in the Despike tab in the Property Panel or reload the data.

Notes:

The Despike button is disabled if the data has been loaded or if there is no entry in the Despike table.

After having set up the Despike table with one or more columns to despike, loading or reloading data is despiked automatically, thus there is no need to go back to this window and manually click the Despike button.

Normal score-transformed variography is despiked automatically.