Financial issues: loans, mortgage, investment

Prices often form patterns that can be evaluated using probability methods, or simply viewed in much the same way as a frequency distribution, or histogram. Because the concepts are sound, but the statistical analysis is often difficult because of limited amounts of data or changing conditions, analysts have taken a much more empirical approach toward studying price distributions. The following section will look at some innovative ways to look at price distributions and how they are interpreted into trading opportunities.

Bollinger bands are a very popular application of price distributions. They do not detrend price, but calculate the standard desiation of prices over a period of 20 days and form a band of 2 standard deviations around the trendline. It is common for traders to vary both the period and the number of standard deviations used to construct the band. Once calculated, Bollinger bands can be displayed on any price chart and used to generate buy and sell signals, much the same as any other channel breakout system. Using a smaller Bollinger band, for example, 1 standard deviation, will give many more signals than using one of 3 standard deviations. At the same time, a band of 3 standard deviations translates into risk that is 3 times greater than 1 standard deviation. Signals produced with a larger band tend to be more reliable, but have greater risk.