mladen wrote: Tue May 09, 2017 9:27 pm The usual issue with Bollinger bands is the following :
_____________________________
Bollinger bands is defined as SMA +- standard deviations
Now - what is less known, is that standard deviations are calculated using sma too. So, they fit naturally (the sma and standard deviations) in the Bollinger bands. But as soon as the sma is replaced for some other average type it becomes broken (that connection) and in some cases (like the proposed tema) I am not sure that a deviation of tema can be calculated at all. It is not just replacing the average for deviation calculations - that would be a "hit-and-run" method - since that will not reflect the nature of the average
_____________________________
To cut it short : some bands that are calculated as some average +- standar5d deviations are easy. But if that will be deviations of that exact average - that is quite a different story