Standard Error Bands by Jon Anderson
Here's a new technique for placing trading bands around the price action of your favorite market. This technique shows how to recognize low-volatility situations and aid in forecasting trend direction.
There was a time when technical analysis consisted of hand-sketched trendlines and bar charts drawn on graph paper, but today, using powerful computers, technical analysis has evolved to include advanced oscillators and sophisticated indicators. Due to software limitations, however, statistically based indicators have been difficult to apply in security analysis, even though statistics, when interpreted properly, can describe many relationships and lead to estimating probable future outcomes. This very point led me to research statistical relationships to find profitable buy and sell signals using statistical tools. The standard error band is such a tool.
One popular technical indicator, Bollinger bands, is a very useful statistical tool. Bollinger bands indicate
areas of high and low volatility around the mean or average price. Bollinger bands consist of three bands, with the computation of the middle band a 20-period moving average of the closing price, and then adding two standard deviations of the closing price to the moving average to form the upper band. The lower band is constructed by subtracting two standard deviations of the closing price from the 20-period moving average. When the bands are wide, the market is volatile, and the narrowing bands indicate that volatility is receding.
While this volatility measurement is of value, the Bollinger band by itself does not indicate the degree or
direction of the market trend. And because most technicians attempt to trade with the trend and not against it, it became obvious to me that a statistically based indicator showing the trend and the volatility around that trend would be necessary. From this observation, I created standard error bands.