Stochastic Learning in Simple Language

Stochastic concerns phenomena which, although partially due to chance, are subject to statistical analysis.

In finance, stochastic is used to identify overbought areas and oversold areas.

Stochastic makes it possible to make predictions about the future behavior of financial assets ( stocks , currencies , etc.) in the form of percentages.

Stochasticd is a stock market indicator developed by trader George C. Lane (1921-2004). Used in technical analysis , the stochastic indicator makes it possible to compare the level of quotation of a security with its past performances over a given period (generally 14 days) in order to “predict” its behavior.

This involves comparing the closing position with the extremes of the day and highlighting a relationship between the closing prices and the maximum or minimum range (gap) of the security during this period of time.

From there, it is theoretically possible to identify overbought or oversold areas, which will allow a trader to place orders at the right time and maximize their gains.

Stochastics is part of the category of oscillators whose purpose is to identify trend reversals. In this, it is close to the RSI.

Reminder : the RSI (Relative strength index) is a financial indicator that both measures the strength of a trend and determines whether the market is overbought or oversold.

The basic principle used by the stochastic oscillator is that of “ momentum ”: the price of an action can be assimilated to a movement which, depending on its strength, will take more or less time to be modified.

In practice, it is composed of two curves limited between 0 and 100% and between which it oscillates.

Graphically, the stochastic is calculated with two variables denoted “% K” and “% D”.

In short, it should be remembered that:

  • the “% K” line indicates whether the current price is higher or lower compared to the lowest price observed over a given period;
  • The “%D” line, corresponds to a 3-day exponential average of the “%K” line.

As the indicator fluctuates between 0 and 100%, there is an overbought zone and an oversold zone:

  • When the oscillator moves between 70 or 80 and 100%, the stock is in the overbought zone: the probability that it will continue to rise is high, so it's time to sell.
  • When the oscillator is between 0 and 20 or 30%, the situation is reversed: the stock is in an oversold zone. The price is too low compared to its history, so now is the time to buy it.

In summary:

  • When %D is above 70 or 80% but crosses this overbought zone on the downside, it is considered a sell signal.
  • When %D is oversold, below 20 or 30% and then crosses this threshold upwards, it is considered to be a buy signal.
  • When %K exceeds %D on the upside, it is considered a buy signal.
  • When %K crosses %D on the downside, it is considered a sell signal.

The stochastic is a useful indicator to confirm (or invalidate) the trend anticipation that a shareholder can anticipate according to certain technical indicators. However, this indicator is not sufficient to make an investment decision on its sole basis.

The Stochastic thus shows the relative position of the price located in a historical gap. Let's study it, still with the title IBM to allow you a comparison with the indicators of previous courses.

In addition to the courses in green, we essentially see a very dancing blue curve. This is Stochastic encrypted as it uses a 14 day comparison period. In addition, it is smoothed over 3 days to avoid a large number of flash spikes that would interfere with readability. This Stochastic curve is called the %K .

Two red horizontal lines have also been drawn. One marks level 80 of the stochastic, the other represents level 20 . Their usefulness constitutes the first rule of use: When the Stochastic crosses below the line 20 , it is said to be oversold . It simply means that the price is too low compared to its 14-day history. The immediate consequence would be buying because the stochastic then shows that this is a cheap security.

In fact, this is not entirely true, because if the title is low compared to its recent history, there is no evidence that it cannot fall further. One could even say that one then buys a value in a state of decline. It is therefore necessary to wait patiently for the value to stabilize a little and consider changing direction. It is therefore much better to buy when the Stochastic comes to cut in the other direction its line 20 .


Thus, the black arrows numbered 5 and 6 show these intervention points .

The reverse situation is also valid. Look at this insistence of the stochastic at point B on the right: The stochastic keeps breaking out of the 14-day history, remaining well above 80: it is therefore appropriate to maintain the position bought. Then, the stochastic collapses and cuts the 80 line thus warning that it is time to take its profit, consequent in this case.

To be able to enact the second rule for using Stochastic, we add a red dotted signal line called %D . It is calculated here by a 14-day average of the Stochastic %K. There are other possible calculations for this %D but the method presented here is the most standard.

As with the MACD, %D is a signal line : it is neither more nor less than an arithmetic moving average of the Stochastic, calculated here over 14 days. We will buy when the %K Stochastic crosses its %D signal line higher . This is the case at points 1, 2 and also at point 3. This last signal allows us to catch our famous final rise already explored in previous lessons. On the other hand, this new sensitivity will often require confirmation by another indicator (macd, bollingers) or an analysis on a different level (trend, etc.). If we rely only on stochastics, then it is good to combine at least rules 1 and 2, ie the exit state of the overbought and the cut of the signal line.

Note in passing a remark by Martin Pring in his book “ Momentum ”: If the crossing is done in the same direction, that is to say that %D cuts %K by catching up with it, the signal has more reliability. This is precisely the case of point 1.

  • wiki/technical-indicators/learn-stochastic-in-simple-language.txt
  • Last modified: 2022/06/30 03:38
  • by admin