Reducing Data Nonnormality with Box-Cox TransformationΒΆ

Non-Causal Step

This step applies the Box-Cox Transformation algorithm to make non-normally distributed data normal.

Input Parameters

  1. Non-normally time-series data

Output Parameters

  1. Time-series data that has approximately normal distribution

Workflow

../../../../_images/workflow78.svg

Algorithm

Box-Cox Transformation

References

    1. Klevecka and J. Lelis, Pre-Processing of Input Data of Neural Networks: The Case of Forecasting Telecommunication Network Traffic, Telektronikk, vol. 3/4, pp. 168-178, 2008.