The Financial Forecast Center™
Independent. Objective. Accurate.
The standard deviation is a concept from statistics and is a measure of the amount of variation (or deviation) that might be expected between the actual indicator value and the forecast value. The standard deviation is given in the same units as the indicator. As an example, the retail sales forecasts are given in U.S. dollars and, thus, the standard deviation is also in U.S. dollars.
Given a forecast value and a standard deviation, the possible range of actual values can be found. From statistics, there is a 68% chance that the actual value will be either one standard deviation above OR one standard deviation below the forecast value, or +/- 1 standard deviation. It also works out that there is a 95% chance the actual value will be within +/- 2 standard deviations and there is a 99.7% chance the actual value will be within +/- 3 standard deviations. Statistics also says there is always some small chance the actual value can be any number of standard deviations from the forecast value, but usually the actual value will be within +/- 4 standard deviations of the forecasted value.
Thus the standard deviation is a very concise and powerful way of conveying the amount of uncertainty in a forecast. The smaller the standard deviation, the less the uncertainty.
As an example, lets say the Dow Jones Industrial Average is forecast at 10000 points and the standard deviation is 200 points. From this the following can be found:
I. There is a 68% chance the Dow will fall between 9800 and 10200.
II. There is a 95% chance the Dow will fall between 9600 and 10400, and
III. There is a 99.7% chance the Dow will fall between 9400 and 10600.
IV. There is some very small chance (about 3 in one million) the Dow will fall between 8800 and 11200
Popular Forecasts