@CavaliereVerde i find Ehler's approach dubious, on par with the likes of John Hurst of Hurst Cycles fame, and here is why:
Ehler comes from an electrical engineering background and for whatever arbitrary reason assumes that the price action of financial markets traces out over time a signal path similar to electromagnetic phenomena as described by physics and thus uses the math from electrical engineering to model market behavior and extract some predictive value after filtering market price data through his indicators.
just like Bachelier did by assuming that economic phenomena could somehow be described by the mathematics of physics as if by some improbable and amazing chance economic phenomena shared so much with physical phenomena that it was scientifically sound to borrow heavily from the mathematical apparatus of physics to talk about another class of phenomena that were totally unrelated. well, it was not sound and the result has plagued economics ever since. this author explains it way better than i could: 'The Origin of Wealth: The Radical Remaking of Economics and What it Means for Business and Society' by Eric D Beinhocker, who is Senior Fellow at the McKinsey Global Institute. Fortune magazine named him a Business Leader of the Next Century, and his writings on business and economics have appeared in a variety of publications, including the Financial Times.
just as Bachelier was a huge assumption, so his Ehler's, insofar as it has no scientific basis. Benoît Mandelbrot demonstrated it in his research summarized in his book for the layman: 'The (Mis)behavior of Markets'.
Mathematical superstar and inventor of fractal geometry, Benoit Mandelbrot, has spent the past forty years studying the underlying mathematics of space and natural patterns. What many of his followers ...
probably this is not the last word on the subject, otherwise how to explain the insane success of Virtu and Renaissance Technologies? nonetheless, there are many golden hints in Mandelbrot's book, not the least of which is that price follows not a Gaussian distribution, but a Levy distribution.
finally, i am going to conclude by mentioning these additional resources for those who aspire to become quants:
https://www.quantstart.com/ebooks and to build your AI engine for trading you could use neural networks either in Python with Theano: http://www.wildml.com/2015/09/speeding-up-your-neural-network-with-theano-and-the-gpu/ and here: http://deeplearning.net/software/theano/index.html
or an easier approach to neural nets here with Lua and Torch:
now try doing this in MQL!