Market Timing With Moving Averages: Anatomy And Performance Of Trading Rules
University of Agder – Faculty of Economics
March 25, 2015
In this paper, we contribute to the literature in two important ways. The first contribution is to demonstrate the anatomy of market timing rules with moving averages. Our analysis offers a broad and clear perspective on the relationship between different rules and reveals that all technical trading indicators considered in this paper are computed in the same general manner. In particular, the computation of every technical trading indicator can be equivalently interpreted as the computation of the weighted moving average of price changes. The second contribution of this paper is to perform the longest out-of-sample testing of a set of trading rules. The trading rules in this set are selected to have clearly distinct weighting schemes. We report the detailed historical performance of the trading rules over the period from 1870 to 2010 and debunk several myths and common beliefs about market timing with moving averages.
Market Timing With Moving Averages: Anatomy And Performance Of Trading Rules – Introduction
Technical analysis represents a methodology of forecasting the future price movements through the study of past price data and uncovering some recurrent regularities, or patterns, in price dynamics. One of the fundamental principles of technical analysis is that prices move in trends. Analysts firmly believe that these trends can be identified in a timely manner to generate prots and limit losses. Market timing is an active trading strategy that implements this idea in practice. Specifically, this strategy is based on switching between the market and the cash depending on whether the prices trend upward or downward. A moving average is one of the oldest and most popular tools used in technical analysis for detecting a trend.
The great controversy about technical analysis is over whether it is scientific or non-scientific. One the one hand, technical analysis has been extensively used by traders for over a century and the majority of active traders strongly believe in market timing. On the other hand, academics had long been skeptical about the usefulness of technical analysis. Yet the academics’ attitude towards the technical analysis is gradually changing. The findings in the papers on technical analysis of financial markets, published in prominent academic journals (examples are Brock, Lakonishok, and LeBaron (1992), Sullivan, Timmermann, and White (1999), Lo, Mamaysky, and Wang (2000), Okunev and White (2003), and Moskowitz, Ooi, and Pedersen (2012)), suggest that one should not dismiss the value of technical analysis. Recently we have witnessed a constantly increasing interest in technical analysis from both the practitioners and academics alike (see Park and Irwin (2007)). This interest developed because over the course of the last 15 years, especially over the decade of 2000s, many technical trading rules outperformed the market by a large margin.
However, despite a series of publications in academic journals, modern technical analysis is still largely based on superstitions and beliefs. Consequently, modern technical analysis remains art rather than science. The situation with market timing is as follows. There have been proposed many technical trading rules based on moving averages of prices calculated on a fixed size data window (called the “lookback” period). The main examples are: the momentum rule, the price-minus-moving-average rule, the change-of-direction rule, and the double-crossover method. In addition, there are several popular types of moving averages: simple (or equally-weighted) moving average, linearly-weighted moving average, exponentially-weighed moving average, etc. As a result, there exists a large number of potential combinations of trading rules with moving average weighting schemes. One of the controversies about market timing is over which trading rule in combination with which weighing scheme produces the best performance. The situation is further complicated because in order to compute a moving average one must define the length of the lookback period. Again, there is a big controversy about the length of the optimal lookback period. Nevertheless, one can easily note that technical traders do share a few common beliefs and myths. They are as follows. First, one can easily beat the market using some technical trading rules. Second, in the computation of a moving average one has to overweight the most recent prices because they contain more relevant information on the future direction of the price than earlier prices. Finally, in each trading rule there exits some specific, time-invariant, length of the lookback period that produces the best performance.
See full PDF below.