The advantage to plotting co variance as a time series is that it will show you how of a trailing period changes. Therefore and other methods of analysis such as Fibonacci retracements could be applied in order to generate targets.
For the purpose of this indicator I have the mean using a derived from . This makes this measurement of co-variance more sensitive to changes in , likewise are more representative a change in , thus giving this indicator a "leading aspect".
//Moving Covariance by Rashad study(title="Moving Covariance", shorttitle="MCV", overlay=false) src = vwap, len = input(30, minval=1, title="Length") mean = vwma(src, len) stdev = stdev(src, len) covariance = (stdev/mean)*100 plot(covariance, title = "moving covairance", style=line, linewidth = 2, color = red)