Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Predicting a crossover or moving average for next bar
Basically the idea is to mathematically calculate a potential prediction for a moving average
""A couple articles in the February 2007 TAS&C caught me eye. One of them talked about a method to attempt to predict a simple moving average crossover the bar before it happens. Basically, they figure out what the price would have to be in the next bar, in order to make the MAs cross. Then, they compare the needed price to the current price. If the MAs would cross even if the stock remained unchanged in the next bar, they predict that it will happen.""
I think that it would be really useful when programmed with LinReg/HMA/Ehlers to anticipate where the price is moving: a useful filter or point of S/R.
Have a look: very interesting to see if anyone possibly already has this coded or could .. the math is in the blog post.
This topic probably deserves its own thread, but I've been wanting to do this for a long time. There are numerous times (especially on a ranged bar chart) where it can be hugely beneficial to know the outcome of an indicator based on if the bar were to close High or Low (the only two possibilities).
Unfortunately, Richard is way smarter than me and I think there is little to no hope I could understand the math involved. So all I can do is hope that someone else writes the code in a way where I can port it to different moving average types.
But thanks very much for the link, I'm glad that others have considered this and found value in it!
I think it would be go to have a thread to discuss possibilities in these kinds of indicators. From my knowledge, there are several approaches to making a predictive indicator:
- linear analysis like some kind of extrapolation technique
- probabilistic analysis - monte carlo?
- non linear applications - neural net, etc.
- some other exotic means
At least we can toss some ideas around.
Regards,
-C
“Strategy without tactics is the slowest route to victory. Tactics without strategy is the noise before defeat.” - Sun Tzu
If you feel this would be better served on its own thread please feel free to move the related posts there ... Im not smart enough to accomplish that with no caffein in my bloodstream ... lol
... which to me makes it the most interesting case. There are times when your price indicators basically have no choice but to go somewhere within n-bars, because there is a strict bound on where price itself can go within n-bars. I think this is an under-utilized phenomenon in TA.
For the more general case, yes, you can do linear or nonlinear extrapolation, or inject energy into the system, or any number of other techniques... but be aware that this is what your "fancy" MA types already do (that's what makes them fancy!), so depending on what you are using you are likely just turning up the gain on their approach, and you won't be happy with the results.
Strangely enough, I've found that it's pretty hard to beat just assuming that price doesn't move anywhere in the next few bars (at least in terms of total squared error). Because... it usually doesn't! Price sits still and sits still, and then explodes to its next plateau in relatively few bars. As always, it pays to tailor your approach to the problem at hand... since if you are trying to identify the explosions it won't do at all to assume price never moves
[P.S. I don't mind being called Gary as long as you keep linking to my site ]
and I couldn't resist the analogy ... quote from the above link :"" and note that it zooms in quadratically faster horizontally than vertically."" just like a lot of my trading experiences
