Gunn, Trading Regime Analysis

Murray Gunn addresses what is one of the most difficult problems, if not the most difficult, for traders—how to know when a rangebound market is about to become a trending market and when a trending market is about to become rangebound. Trading Regime Analysis: The Probability of Volatility (Wiley, 2009) is an almost avuncular approach to this problem. That is, imagine sitting down with your favorite uncle as he tells tales—some engrossing, some humorous, some familiar—and then offers words of wisdom, making sure you understand that these words are written not in stone but in sand. He also confides that your aunt sees otherwise. And, oh yes, your uncle is a Scotsman and a currency investment manager.

Gunn assumes that we know trends and rangebound markets when we see them (at least once they’re reasonably well established) and that we don’t need precise definitions. Moreover, he assumes that we have decent strategies for trading each kind of market. The problem is how to know when these trading regimes are shifting and hence when to start shifting strategies. Although Gunn acknowledges that a great deal of quantitative research has been done in this area and even devotes a fascinating chapter to the work of the quants at RBS, he argues that technical analysis is not only up to the job but may even be better because it can not only identify the appearance of a shift but can anticipate the market’s direction.

Gunn’s primary tool is a simpler form of historical volatility. Historical volatility measures the standard deviation of the changes in price over a certain time period. By contrast, Gunn chooses to measure the standard deviation of prices themselves, the dispersion of market prices over a certain time period. “The speed of the changes in the price of a market is not as important as the distribution of those prices when we want to analyse the current trading regime and the potential future trading regime.” (p. 60) That is, he wants to look at the expansion and contraction of price ranges, not the rate of change in the prices. His indicator is similar to the Bollinger Band Width, though the latter describes the width of the standard deviation bands from a moving average rather than simply the standard deviation of the closing price itself. (Gunn then applies a moving average to the standard deviation and analyzes the width between the two.)

Gunn doesn’t follow up on this idea in a rigorous way; instead, he discusses various well-established technical indicators and charting techniques that can be part of a trading regime analysis. He admits that all of these work some of the time, that none works all of the time, but that by mixing and matching the trading regime analyst should be able to determine the change of regime quickly enough to capture the meat of a trending move and avoid being chopped to death when the market trades sideways. Gunn’s approach is eclectic and “soft.” He’s essentially searching for “a good feel.” As such, it’s somewhat frustrating. But then markets are frustrating as well and don’t succumb easily, if ever, to mathematical analysis. Or, as Gunn would say, in one of his favorite expressions, they do and they don’t. They yield a little here, a little there, but those darned traders and investors with all their quirks keep messing up the models.