AI-powered Trading System

[trade4win]

Can someone recommend a proven AI-powered tool? I extensively tried TradingSolutions for a couple of months but haven't managed to create a profitable strategy based on neural nets. I'm thinking about purchasing Alyuda Tradecision Pro (www.tradecision.com) or NeuroShell Trader Pro (www.neuroshell.com). What would be your recommendations? Thank you.
Regards

[theplumber]

Trade signals or models? Two very different elements to consider.

[spike500]

Re: AI-powered Trading System

Originally posted by trade4win
Can someone recommend a proven AI-powered tool? I extensively tried TradingSolutions for a couple of months but haven't managed to create a profitable strategy based on neural nets. I'm thinking about purchasing Alyuda Tradecision Pro (www.tradecision.com) or NeuroShell Trader Pro (www.neuroshell.com). What would be your recommendations? Thank you.
Regards

------------------------------------------------------------------------------------

Don't buy ANY blackbox at all, this will be the best investment you ever made.

“The Road to Trading Wealth Cannot Be Automated -- The Myth of Mechanical Systems Dispelled”

Imagine a salesman called you up on the phone and told you that for just a few thousand dollars or so, he could give you a software program or computer chip that would allow you to automatically play the piano (or any instrument) in your home like a professional musician and you would never have to practice again.

You could amaze all your friends and family with your newfound skills. You would probably laugh the man off the phone. But if the same man offered to sell you a mechanical trading system that would make you wealthy without having to do any work, would you still laugh?

“Many people don’t and they are eager to purchase such a system, but...”

…traders who believe in the myth that you can just hand over your hard earned capital to a computer program and sit back and collect checks and build wealth are not dealing with reality. They are driven by greed and laziness.

They want to make the money successful traders do, yet they do not want to put in the hard work necessary to develop the skills needed to become and stay successful.

Just think about it for a minute. If this were so, why don’t the best and most successful traders just go out and buy the best one or two systems out there, hand them over a broker and just have them execute the systems while they lay on their beach chair in Tahiti and collect a never ending stream of checks from their trading account?

You wonder, because you see they are still coming in early and staying late and working their tails off to make their trading work. It’s just not that simple. If it were, there would be no money left for all of us who wish to trade and make money. This is because the few or everyone who just bought the best system out there would hoard all the money and ruin the game.

“In Trading, You See, There Have To Be Losers For The Winners To Earn Money.”

Complex human activities cannot be automated. Driving your car, flying an airplane, playing an instrument are just some of the examples of activities that cannot be programmed into a computer and work all the time.

Some of the elements of these activities can be programmed for sure, but the bottom line is that a human needs to apply their skills and talents to make the whole thing work when unexpected situations, such as when the Fed announces interest rate decisions, the government announces they are going to stop issuing 30 yr T-Bonds etc. arise.

Trading is no different.

As long as there has been trading and investing, there have been trading programs sold to the public. Most of these systems are sold for $3000 to $10,000 and up. I find it hard to believe that there has been a system sold to the public, which has produced consistent profits for more than one, or two years before the account was wiped out or that the individual just hung it up. Most of these systems sold are scams.

[Gavishti]

I agree that there are no money machines available for purchase. Although many are advertised. There are, however, automated trading programs that do work. Some trade many millions of dollars. Each one has been developed through much hard work. To develop a successful system you first need to know how to trade successfully on your own. Then you must be able to instruct the computer how to do what you do. Few individuals possess the skill to do both. Given enough time and determination most people should be able to obtain both skills. These skills are not developed in a few weeks. It will take years.

[theplumber]

Re: Re: AI-powered Trading System

Your response doesn't make any sense to his question, that you quoted no less. Re-read what he said, "couldn't make a profitable STRATEGY". That's what the products he mentions do, they take stochastics and say a moving average and try to generate a trade signal (that is his personal discretionary decision for the criteria) that's more profitable than if he tried to recognise it on his own staring at a screen all day. This by definition isn't a trade system, AGET is a trade system. Do you know how exactly all the tools in AGET work? Then it can be said it's a form of AI that give trade signal that you don't know how it works, which by definition is a black box. So I guess your saying don't get AGET right? POS add-on black box and won't make you any money right? The products he mentions are neural networks, where you have to decide what inputs to and how the data is trained, 100% control by you. Don't understand NN's? Program one your self and find out that it isn't a black box. Wave59 has a neural network, has technical indicators as inputs. Black box? Get real, only tries to help make the signal come sooner than you recognizing it. I have Statistica's neural net that I bought for biophysics research, and imagine this I can change the imputs to cycles data and make a forecast model for my emini trading. Black box? Ya, one that works for biophysics and the stock market (sarcasm there, sorry).

Originally posted by spike500
------------------------------------------------------------------------------------

Don't buy ANY blackbox at all, this will be the best investment you ever made.

“The Road to Trading Wealth Cannot Be Automated -- The Myth of Mechanical Systems Dispelled”

Imagine a salesman called you up on the phone and told you that for just a few thousand dollars or so, he could give you a software program or computer chip that would allow you to automatically play the piano (or any instrument) in your home like a professional musician and you would never have to practice again.

You could amaze all your friends and family with your newfound skills. You would probably laugh the man off the phone. But if the same man offered to sell you a mechanical trading system that would make you wealthy without having to do any work, would you still laugh?

“Many people don’t and they are eager to purchase such a system, but...”

…traders who believe in the myth that you can just hand over your hard earned capital to a computer program and sit back and collect checks and build wealth are not dealing with reality. They are driven by greed and laziness.

They want to make the money successful traders do, yet they do not want to put in the hard work necessary to develop the skills needed to become and stay successful.

Just think about it for a minute. If this were so, why don’t the best and most successful traders just go out and buy the best one or two systems out there, hand them over a broker and just have them execute the systems while they lay on their beach chair in Tahiti and collect a never ending stream of checks from their trading account?

You wonder, because you see they are still coming in early and staying late and working their tails off to make their trading work. It’s just not that simple. If it were, there would be no money left for all of us who wish to trade and make money. This is because the few or everyone who just bought the best system out there would hoard all the money and ruin the game.

“In Trading, You See, There Have To Be Losers For The Winners To Earn Money.”

Complex human activities cannot be automated. Driving your car, flying an airplane, playing an instrument are just some of the examples of activities that cannot be programmed into a computer and work all the time.

Some of the elements of these activities can be programmed for sure, but the bottom line is that a human needs to apply their skills and talents to make the whole thing work when unexpected situations, such as when the Fed announces interest rate decisions, the government announces they are going to stop issuing 30 yr T-Bonds etc. arise.

Trading is no different.

As long as there has been trading and investing, there have been trading programs sold to the public. Most of these systems are sold for $3000 to $10,000 and up. I find it hard to believe that there has been a system sold to the public, which has produced consistent profits for more than one, or two years before the account was wiped out or that the individual just hung it up. Most of these systems sold are scams.

[spike500]

Plumber,

what i wanted to explain is that 90% of the so called black boxes with huge returns only bring profit to the seller of the box. There may be some systems that work well.
But than i would like to see a track record of at least three years of real trades with returns calculated according the rules of NFA and CFTC.
In the US there is a group called "Club 3000". This group was founded by people who bought black boxes that costed around 3000$ a piece and never where succesfull in trading these black boxes. 3000$ is just no too expensive to make people take the risc to buy.
I have also developped a system. I will never sell it, even not for 50 000$.

I don't exclude that there will be systems that work, but that will be the exeptions that confirm the rules.

My english is not so good, so perhaps i misinterpreted the posting.

[its1111]

Mr. Plumber Wave 59

Plumber

I ran across the wave 59 software the other day and then read your post and you mentioned it. Do you use it, or have you heard it is good, bad? What software do you use or is it your own. I am always interested in trading ideas.
Thanks
Mark

[theplumber]

Re: Mr. Plumber Wave 59

Originally posted by its1111
Plumber

I ran across the wave 59 software the other day and then read your post and you mentioned it. Do you use it, or have you heard it is good, bad? What software do you use or is it your own. I am always interested in trading ideas.
Thanks
Mark

Yes I use it, along with Ensign and Esignals advanced charts. In fact I have access to Bloomberg and CQG but use the above software. I'm not a vendor or service provider but a trader of eminis

[theplumber]

Re: AI-powered Trading System

Originally posted by trade4win
Can someone recommend a proven AI-powered tool? I extensively tried TradingSolutions for a couple of months but haven't managed to create a profitable strategy based on neural nets. I'm thinking about purchasing Alyuda Tradecision Pro (www.tradecision.com) or NeuroShell Trader Pro (www.neuroshell.com). What would be your recommendations? Thank you.
Regards

Are you ever going to answer my question? I've used NeuroSolution, which is the big brother to Tradesolution, and had pretty good success with it. I wanted to make a model using genetic algorithms to train my data vs fuzzy logic and found theirs to be exceptional. As for the data input, first it was time series data alone but later added ephemeris data. This is what I was talking about, it all depends on you and what data you chose to use. There's an idiot on Elite who doesn't trade who will show one url after another about how NN's don't work for financial modeling but the models that do work aren't known, because people are to busy making money with them. I only use models and have found something like Neuroshell, or any other technical indicator NN useless, and I would stay away from Alyuda b/c it is a totally unproven NN with no info on exactly they use for their structure. Most though use neural networks for derivitive arb trades so if that's your thing then this is the wrong forum b/c that would be better served with a custom network design.

[trade4win]

NN soft

Plumbe,r
What you can say about BioComp Profit? Have you used it? Is there good neural nets prediction?

[spike500]

Re: NN soft

Originally posted by trade4win
Plumbe,r
What you can say about BioComp Profit? Have you used it? Is there good neural nets prediction?

You need a proven trackrecord. Otherwise you risk to by a system that might be worthless and expensive.
If you buy a car you have the garantee that what you expect you will get( normally). So ask the same question for a traydingsystem. You buy it to make profit, let them prove that you will make profit. After all that's the publicity all these systems make: you will earn lots of money with their system.

I don't say that there are no systems that work, but check it before you buy. And if it really works the seller will have no problem in showing you the profits it will generate.

If i would be a seller of a supersystem, the first thing i would do is show a trackrecord. After that the sale will be 10 times easier to accomplish, and it would be impossible to deny the results.

[theplumber]

Re: Re: NN soft

Originally posted by spike500
You need a proven trackrecord. Otherwise you risk to by a system that might be worthless and expensive.
If you buy a car you have the garantee that what you expect you will get( normally). So ask the same question for a traydingsystem. You buy it to make profit, let them prove that you will make profit. After all that's the publicity all these systems make: you will earn lots of money with their system.

I don't say that there are no systems that work, but check it before you buy. And if it really works the seller will have no problem in showing you the profits it will generate.

If i would be a seller of a supersystem, the first thing i would do is show a trackrecord. After that the sale will be 10 times easier to accomplish, and it would be impossible to deny the results.

I agree proven track records would help but the real problem is the software takes so much time to really use it that most don't have the patience to stick with it long enough to get any results that can be good or bad. What I mean is most play with the software for a few days and come to the conclusion it's worthless. What is needed for these types of software is weeks of time. Put it this way, it took me 2 years to come up with a model using Statistica's NN using single spectrum and cross spectrum time series, only now you can buy similar results with Divergence software's cyclepack.

[theplumber]

Re: NN soft

Originally posted by trade4win
Plumbe,r
What you can say about BioComp Profit? Have you used it? Is there good neural nets prediction?

Sorry, I have not used this software. These types of software are very time consuming so trying 3 for me is my limit.

[spike500]

Plumber

I bought 10 years ago an AI software program called Brainmaker.

What i expirienced about AI is the following (maybe this info has become too aged by now):

you buy a tool, not a ready to easy trading system.

you have to feed the program with the right information, that's not so easy.

you need quite a bit of knowledge in mathematical formulas.

it mainly depends on the person if the program will be succesfull, not on the program itself.


I would compare it with any other software program, some people can do wonders with it, others can't even program the most elementary things. I think that's the main problem.

[theplumber]

Re: Plumber

Originally posted by spike500
I bought 10 years ago an AI software program called Brainmaker.

What i expirienced about AI is the following (maybe this info has become too aged by now):

you buy a tool, not a ready to easy trading system.

you have to feed the program with the right information, that's not so easy.

you need quite a bit of knowledge in mathematical formulas.

it mainly depends on the person if the program will be succesfull, not on the program itself.


I would compare it with any other software program, some people can do wonders with it, others can't even program the most elementary things. I think that's the main problem.

Exactly right my friend.