Missing Values

The Missing Values Transform enables you to specify how missing values are treated.

A data value can be missing for a variety of reasons:

• If the data value was not measured, that is, it has a Null value.

• If the data value was not answered.

• If the data value was unknown.

• If the data value was lost.

Data Mining algorithms vary in the way they treat missing values:

• Ignore the missing values, and then omit any records that contain missing values.

• Replace missing values with the mode or mean.

• Infer missing values from existing values.