This is an updated version to the Currency Strength Measurer I created before. I'm not updating that script due to differences in the features and in the idea. The original script was a lagging indicator, I hope that this one is more of a leading indicator, the idea is the same. This indicator has way fewer inputs and it's more lightweight. Try it and compare it with the original SCSM and choose whichever suits your strategy best.
I'm open to comments or ideas on how to upgrade it.
@DBonacich gave me a good suggestion on setups using this indicator. The idea is to keep the smoothing to a minimum in larger timeframes and increase the smoothing in lower timeframes, hope it helps everyone. Please post a comment if you find other good or interesting setups.
Rise_of_Skywalker
⋅
Awesome Script. Why I can't see the label on the right hand side?
Deep-Peat-Shark
⋅
Nice job! What is in short the math behind each currency strength? I see for example that you used exponential moving averages and at the end it looks like a weighted average. Can you explain your reasoning behind the choices of EMA instead of MA and the weighting? Thanks
EilderJorgeGarcia
⋅
@ArnaudKleinveld, Well, I honestly came up with the numbers myself using some of my math background, it's public so that anyone can change it to their own tastes, my idea was the following:
If I get the average of each currency value by either their MA or their EMA and then use a similar base to all currencies (to avoid giving more weight or less weight to each one, instead leaving the EMA or SMA to decide the current weight of each one) I should get more or less an approximation of the current value of a currency, that was the original SCSM, as for why EMA instead of MA, EMA gives more weight to recent values, so it's more responsive...
I used the first version for a while but I noticed it was too slow, too lagging, then I came up with the idea of using the same formula, but add momentum to the mix, so that it becomes more of a leading indicator... after testing some formulas for a while I came up with the idea to use momentum in three steps, giving more weight to the recent steps, then smooth it with an EMA... and it worked, it became a lot more responsive and more of a leading indicator... after thinking a bit more, it made sense, after all the idea to predicting future prices lies in the previous movement, price action as we call it.
The best way to upgrade it would be to assign weights directly to each currency, I.E: Euro and USD are very opposite, while EUR and GBP tend to go more or less together... the algorithm actually reflects that better than I thought, but perhaps manually assigning weights instead of dinamically could be better.