Refining Automatically Generated Confidence Regions for Restricting Outliers in Economic Data

The paper discusses the refining of a methodology for the automatic generation of confidence region, conditional on the joint distribution of data variables. The study is focused on analyzing the reasons for distortion of the confidence region shape due to the higher degree polynomial, used for the automatic generation and its inability to distinguish between the interchangeability of the 0-degree and 360-degree angles at two ends of the range 0 to 360.