Trading Systems and Methods, 6th Edition

CHAPTER 6Regression Analysis

Regression analysis is a way of measuring the relationship between two or more sets of data, or just price and time. A stock analyst might want to know how the price of Barrick Gold Corporation (ABX) changes with the price of physical gold. An economist might want to know how the value of the U.S. dollar is dependent on interest rates, inflation, and the trade balance. A hedger or arbitrageur could use the results to establish the relative fair value of two related products, such as palm oil and soybean oil, in order to select the cheaper product or to profit from price distortions; or, as an investor you might simply want to find the strongest stock in the banking sector. A straight line fit through a series of prices is also a way of drawing a trendline. Regression analysis is a valuable tool for traders.

COMPONENTS OF A TIME SERIES

Regression analysis is often used to identify the main component of a time series, the trend. With some variation, it can also be used to isolate the seasonal (or secular trend) and cyclic components. These three factors are present in all commodity price data as well as many stocks. The part of the data that cannot be explained by these three elements is considered random, or unaccountable price movement.

Trends are the basis of many trading systems. Long-term trends can be related to economic factors, such as changing interest rates, inflation, shifts in the value of the U.S. dollar due to the balance of trade, ...