Configuring: Correlation

Select whether the row number (index base) or the timestamp (time base) is to be used to calculate the correlation between the selected fields. The resulting dataset will contain the correlation matrix for these fields.

Worked example: Correlation

Steps to configure

Correlation base selection: Select the correlation calculation base to be used.
Correlation field selection: Select the fields for correlation.
Execute: A correlation matrix is created, indicating the relationship between the selected fields.

Step 1: Correlation base selection page

Select whether the row number (index base) or the timestamp (time base) is to be used to calculate the correlation between the selected fields.

NOTE: This operation does not support milliseconds. First use the Resample operation to remove milliseconds present in your data source.

Index based is calculated over the number of rows in the dataset.
Time based correlation calculations take into account the amount of time a data point remains at a timestamp value. The difference between timestamp values is used for the calculation.
When you select time based, you need to select the timestamp field to be used. If there is no timestamp field in your dataset, this option will be unavailable.
Default: index based is selected.
Minimum configuration: either index or time based should be selected. If time based is selected a timestamp field must also be selected.

Step 2: Correlation field selection page

Select the fields that should be correlated.

Only the double and integer fields will be available for selection.
Rows in the dataset fields that contain null or empty values will not be included in the correlation calculation. Only rows that contain valid values in the specified fields will be used for the correlation calculations. A value of zero is a valid value.
Default: all double and integer fields are selected.
Minimum configuration: two or more fields need to be selected.

Step 3: Execute

The newly created dataset will be the correlation matrix, showing the relationship between the selected fields of the dataset.