On the Design of Seasonal Adjustment Methods using Linear Programming Techniques

Abstract

The objectives of the seasonal adjustment of monthly economic time series are discussed. The design of seasonal adjustment methods of the moving average type is formulated in terms of a linear programming problem. A moving average is designed, using this technique, to seasonally adjust current data. The designed moving average is compared to the X-11 current data moving average in both the time and the frequency domains.

Key Words:

• Seasonality
• Seasonal adjustment
• Time series
• Linear programming
• Filter design
• Discrete Fourier transforms