New Trading Systems and Methods, Fourth Edition

CHAPTER 11

Cycle Analysis

The cycle is another basic element of price movement, along with the trend and seasonality, but it is more difficult to evaluate; therefore, it is often overlooked. Cycles come in many forms—seasonality, production startup and shutdown, inventory or stocks, behavioral, and astronomical. Seasonality is a special case of a calendar or annual cycle. Seasonality was covered in the previous chapter and its special features are not considered here. Some of the cycles are clearly periodic, having regular intervals between peaks and valleys; others are more uniform in their amplitude (height) but irregular in period. The most definitive and regular cycle remains the seasonal, which is determined by periodic physical phenomena, the changing of the year.

This chapter discusses the major commodity cycles that result from business decisions, government programs, long-term market characteristics and phenomena. Short-term cycles are usually attributed to behavior and will be covered in Chapter 15. There are a few important ways to find the cycle, the most common being trigonometric curve fitting and Fourier (spectral) analysis. Both require a computer and will be explained in the following sections. John Ehlers introduced Maximum Entropy Spectral Analysis (MESA), which finds price cycles based on small amounts of data, at the same time avoiding some of the problems inherent in other methods. Examples of solutions will be included in the explanation of the methods ...