the hAnDbook of technIcAl AnAlySISReplacing C with current RSI value RSIC, Ln with the lowest RSI value of last
n periods RSILn, and Hn with the highest RSI value of last n periods RSIHn, we get:
Stoch RSI moving average of RSI= ( C−RSILn) /(RSIHn−RSILn)The StochRSI is a bounded oscillator, due to the nature of the normalizing
equation above. See Figure 8.17. It shows the stochastic, StochRSI, and RSI win-
dow oscillators, in descending order on the charts. The defaulted lookback period
for the StochRSI is 14 bars. Notice that the StochRSI provides faster and more
responsive signal action as compared to the original RSI.
Data Drop‐Off effect
The drop‐off effect is fairly obvious when an oscillator or overlay drops off an old
data point and substitutes it for a new one, preserving the length of the lookback
period. Let us use a simple moving average to illustrate the effect. Assume that we
create a seven‐period simple moving average of the following prices:
$7, $4, $3, $3, $2, $4, $5
The average would be $28/7 = $4. Assume that we now move on to the next
data point, dropping the last price of $7 and adding a new data point at $4:
$4, $3, $3, $2, $4, $5, $4
The new average will now be $3.57. The percentage change is ($4 − $3.75)/$4
× 100% = 10.75%. If we now replace the original last price of $7 with $9, the
fIgure 8.17 Creating the StochRSI from Normalizing the RSI.
Courtesy of Stockcharts.com