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)
a) demonstrate our use of Chaos Hunter as we apply it to an intra-day scalping strategy for the ES;
b) generate interest and possibly gain support from other forum members;
c) collaboration with the objective of designing a better predictive algorithms; better money management & risk rules.
Notes:
a) We trained the network in Chaos Hunter and exported the results into Easy Language, where we built a committee of models; polling conditions etc. When the model starts to degrade in performance, we will either retune it or replace it.
c) BEFORE YOU LOAD THE ELD files, please view the screen shots for the ELD Files. If you have Functions with the same names, importing the file may lead to over-writing.
c) Before you try to demo the strategy, please open the Strategy txt file and read our notes.
d) We encourage forum members to trade this strategy only in simulation mode.
Loading instructions:
1) Import functions. The Locked function is the Haar Wavelet and it will expire on 12/25/10 rendering itself and the strategy unusable. {If we collaborate, we're happy to extend this time frame or we may elect to give you the indicator.}
2) Cut and Paste the strategy into a new strategy. Please use NWA_ES_R3.(Thanks)
3) From the screen shots, please follow the setup with particular attention to the DATE RANGE for the data set. The date range is important because variables are initialized in the strategy by a start bar. First bar = 09/26/2010 18:05 EST.
We hope you enjoy the strategy; we'll endeavor to post more in the future.
The strategy posted here runs in Tradestation; it's optimizable and performs in paper trading without issue.
Yesterday while attempting to optimize the strategy from within MC, we noted an error from the "sqrt" function. We checked the related functions and they appear OK.
We'll try to sort the issue and repost as soon as it's resolved.
Thanks for sharing it. I unfortunately do not have Tradestation so can not load the files.
From my very basic understanding of NNs, you first need to train them using historical information; this requires you to provide the desired result for the training data. Then when you feed them new data, the expectation is that the NN will produce the desired results based on the history.
Just browzing at the code, I did not understand how the system is setup to train the NN. A pointer will be useful.
Thanks for reviewing our posting. We're working with MC right now to resolve the issue .... 2hrs on Team Viewer, ouch.
We trained the network in Chaos Hunter and exported the results to TS, where we built a committee of models; polling conditions etc. When the model starts to degrade in performance, we will either retune it or replace it.
One of the techniques that we are attempting to employ is that we give MC/TS a pallete of say 5 models. Then we toggle these models from 0 to 1 in an Exhaustive optimization creating a consensus for Entry/Exit conditions.
The output would be something like this:
Long Entry - Use Models 1,3,5.
Short Entry - Use Models 1,3,2, 5
Long Exit - Use Models 1,2
Short Exit - Use Models 3, 5.
Each Model has it's strengths and weaknesses. We would use exhaustive optimization to tell us the contribution of each to Net Profit.
The Networks are trained on the dataset that they were built upon, and this tasking is performed by Chaos Hunter. Hope this answers your question and gives some ideas for future work.
I've just noticed some of your files are encrypted (non-source) and you've also set an expiration date. We do not allow encrypted (non-source) attachments on futures.io (formerly BMT) for safety reasons. Please replace the attachment.
Also, can you clarify your intent with regards to the expiration date? If you are intending to sell this tool, I'm afraid you are not permitted to post it here or promote it here. Please clarify.
The purpose is clearly outlined in our posting, right?
"The purpose of this posting is as follows:
a) demonstrate our use of Chaos Hunter as we apply it to an intra-day scalping strategy for the ES;
b) generate interest and possibly gain support from other forum members;
c) collaboration with the objective of designing a better predictive algorithms; better money management & risk rules."
We locked one of the files because we are offering a strategy and that shouldn't mean to us that we need to give away proprietary inputs to the general public. We offered to extend the timeframe; and we offered to give away the indicator to those who make a meaningful contribution.
If we elect to sell some of our work at a later date; it will be fully tested; properly documented and used by us before it is ever distributed or advertised, and any advertising will be done at approved places; with proper clearances; and in a professional way.
So to be clear to forum members and BM - we are not selling a product; we are not advertising a product; the indicator is locked because we are posting a strategy for the purposes outlined several times. If that's an issue, BM kindly remove the posting; forum members please accept our apology, and we will post another strategy in the future with full open source.
{Side note - I have not heard back from MC but I'm confident they are pursuing the issue. They mentioned to me that it would take some time to resolve}.
We never made a live system trade using our own money but today, we elected to go live with an Entry.
Within seconds the strategy jumped short at 1158.25. Having never used strategy generated orders, I wanted to understand the functionality. So, I disabled the strategy after it entered and verified that TS accepted our working stop. After confirmation, I cancelled the stop and placed a bracket order so I could watch the strategy.
I bailed at 1155/3.25 and the strategy bailed at 1153/5.25pts.
Some limited testing but it was fun to watch a computer make a trade.
We pulled together this simple multi-model neural strategy so we could further demonstrate our work. Unlike our previous posting, it's all open source and we verified functionality in MC and TS; there are two separate files.
Please note the initialization instructions in the notes section of the strategy. The ChaosVars on 9/26/10 @ 1805. From within MC you can set the start date 9/26/10 and from within TS 9/27/10.
Over the long haul this strategy looks ugly but you'll notice that it also has good periods. So, how can we turn an ugly strategy into a working strategy? Here's a few ideas if anyone is interested:
a) Because the strategy consists of 2 models, it has 4 x 2 threshold conditions. We could optimize for the best entry/exit combination by toggling the inputs between 0 and 1. For example, model 1 may give the best LE whereas model 2 may only harm the strategy - you get the idea. If you find a good combination, make sure you edit the Entry/Exit flag conditions. You can probably just set them to <=1 before you start; remember there are (4) and after optimization if you find two model conditions work best, you change Entry or Exit flag to take that into account = 2.
b) We could optimize for a stop/profit target in connection with (a) or after we identify the Entry/Exit combinations. Maybe the model will eliminate this requirement.
So, if you have a spare I7 sitting around and some free time, perhaps you can give it a try.
If we botched anything let us know and we'll do our best to correct it.
Attached is a NN Strategy and Indicator; we made it simple so that it could optimized for current data - nothing last for ever.
You can start by optimizing for TRIX length Fast & Slow, and occassionally optimize for threshold values. Once you have the appropriate values, you can modify the indicator inputs accordingly.
We noted that a common indication for losing trades vs winning trades are long dwell above/below the threshold value(see zip file screenshots). This strategy / indicator combination can be used for discretionary trading in conjunction with your favorite tools. We checked it over 5 years and the performance is so-so, but when optimized over shorter durations the walkfoward looks hot.
Give it shot; it's untested in MC but should work without an issue.
This might be a good tool for someone who likes TRIX line trading.