STOCHASTICS
The Stochastic oscillator compares closing prices relative to high-low range over a specific time period.
The Slow Stochastic is derived from the less popular, and far more sensitive, Fast Stochastic, which produces two lines – %K and %D.
The fast stochastic calculations are as follows:
| %K = 100[(C – L10) / ( H10 – L10)] where C = last close, H10 is the highest high in 10 bars and L10 is the lowest low. %D is normally a three-period smoothed average of %K. %D = 100 x (H3/L3) where H3 is the three-period sum of (C- L10) and L3 is the three-period sum of ( H10-L10). |
The slow stochastic takes this %D value and converts into %K and then smoothes once more by adding a new %D which, again, is a three-period average of %K.
Users can choose between the Original and the Simple Algorithms. The formulas for each are:
| Original Algorithm = [Moving Average (Closing Range)] / [Moving Average (Total Range)] Simplified Algorithm = Moving Average (Closing Range/Total Range) |
Closing Range = Close Range Minimum
Total Range = Range Maximum Range Minimum
Generally, the simplified algorithm is more responsive to price movements:
- A stochastic value above 75 indicates an overbought market, while a stochastic below 25 indicates an oversold market.
- The crossing of the %K and %D lines generates buy and sell signals when the stochastic has been overbought or oversold and then moved out of the zone. This is especially important in a strong trend as the stochastic will record extreme levels and give false crossovers.
Stochastic can show divergence, when Slow %K and Slow %D values decline and closing price values increase, or Slow %K and Slow %D values increase and closing price values decrease. The Stochastic values are moving in one direction and the price values are moving in the opposite direction.
The normal setting is 14,3,3 or 10,3,3, but this can be overly sensitive to price movement. Changing the settings to a longer period such as 21, 13,8 will give less frequent crossovers, and the overbought/oversold levels will be more realistic as they will occur less often.