The R-Squared or R2 is a statistical measure that shows the percentage of movements of one time-series that can be explained by movements of another time-series. The R-squared values range from 0 to 1 or 0 to 100 if expressed in percentage. A value of 100 indicates that all movements in one time-series or security are explained by movements by another time-series or security. A value of 0 indicates that that no movements in the first time-series can be explained by movements in the other time-series.
The R-Squared is simply the correlation between two time-series raised to the second power.
Chande & Kroll introduced, in their book 'The new Technical Trader', an indicator that uses the R2 tool. According to Chande & Kroll, the indicator is a lagging indicator that shows the strength of the trend and that the primary use of R2 is as a confirming indicator.
The indicator accepts one parameter, which is the number of periods (bars) used to calculate the correlation. The default value is 14. The time-series produced by this R2 indicator bounces between 0 and 1. You can multiply it by 100 to get a percentage value.
You can also use a moving average to smooth the result:
sma(chande_r2(14) * 100, 14);
In this example, I have used a moving average with a period of 14 bars and I have also multiplied the Chande & Kroll R2 indicator (chande_r2) by 100.