These indicators are all based on a 2004 April issue of Technical Analysis of Stocks & Commodities article by David Sepiashvili. Originally there were only 2 versions developed by the author, but there were some possibilities that were not explored in that article which lead to developing some more variations of those 2 indicators
Some preface : in the attached article the author does a well job explaining what he felt a 2 most important points of Trend Quality indicators calculation. What is not explained there is how the "direction" (or "reversal" as the coders of tradestation, metastock, amibroker ... call it) is determined. It is a difference of 2 emas (7 and 15 ema by default) and since the difference of 2 emas is by definition macd, it turns out that the additional "spice" in these indicators is macd. It does not make these indicators a "yet another macd" indicator, since they are not that, but I felt it necessary to clarify that point too (to avoid some mis-expectations)
There are 2 "flavors" of Trend Quality indicators : the Q indicator (which is in a strict sense the "quality" indicator) and the B indicator (or a "trend-noise balance indicator") Both of these versions are going to be varied since both can be modified in order to show the whole thing a bit differently.
Trend Quality indicator(s) ...
These 2 are the original indicators.
The main property of the original indicators is that they are not "directional. Instead they are showing the "quality" (or strength) of a current trend) so the usage of levels recommended by David Sepiashvili is essential in these, but the direction of the trend must be determined discretionary by the user of the indicators.
Q indicators are oscillators without a known bound. They use predefined levels in order to determine the strength or the quality of the current trend (those levels are described in the attached document)
The B indicator(s) are bound oscillators and they are having a 0-100 bounds. For the purpose of better visibility of a "signal" ("reversal", or "macd trend" values) minimum is not set to 0. Levels of this indicator are also described in the attached document and are (due to the nature of the indicator) different in values than the Q levels.
As it is visible from the example, they really are not "directional" indicators. The reversal (or some kind of a direction) is provided by the signal line that is in fact showing macd trend (which, as we will see, does not have to conform with the Trend Quality indicators trends)
Trend Quality indicator(s) ...
These 2 are "variation 1" on the theme of Trend Quality indicators
In these versions indicators not showing just the quality (strength) of the trend but the direction of it too. As it is immediately obvious, the macd trend and the trend quality trends do not have to conform at all (and in fact are not - it seems that in cases when they are not conforming, Trend Quality indicators are leading when compared to macd, but that is a fact that needs to be tested thoroughly)
These 2 are "variation 2" on the theme of Trend Q
For the sake of making it easier to see what have been done on it, here is a comparison of the regular Q indicator and this "smoothed" version
And the 2 are the last variation on the theme (for now )
These are the smoothed version made directional. All what has been said for the "regular directional" version applies to these too, except that one thing probably needs to be said : the "directional" change is not done by simple mirroring around zero line, so there are case when the directional and the non-directional version do not conform in values, but as far as I see, those case are logical and in my opinion they should stay as is since I feel that it gives them some additional quality. As far as the signals are concerned (zero line crosses), this one is the closest to macd, but the rest (levels crosses and thus trend quality determination is in my opinion the most important in this version and they should be used instead (level not the zero line crosses as signals)
As far as I could see, there is some good potential in all these indicators, but don't take my word for it Let us test them and see if they can help us in our trading
Would it be possible to build a dashboard out of it across multiple pairs and time frames, so you know where to focus your day trades for the day?
MTF capability ?
these are very interesting indicators. Is it possible to add MTF capability ?
Here you go Just one explanation of a deviation to usual multi time frame interpolation. On the picture you see that there is an empty space in histogram when the indicator changes trend. I was thinking if I should interpolate histogram for that part or not, and in the end decided not to do so since this way it is clear that the trend is changing and that the value of the indicator itself in those points is 0
These do look promising. Is it possible to create dynamic versions of the indicators ?
Do you mean dynamic zone versions?
Yes, I mean dynamic zone version of the indicators.