Leveraging Collective Intelligance to predict price
...
Great idea...count me in....
excellent idea, i already joined and passed the word...
Sounds like a a great experiment to me.
One question, how did you arrive at the 1,500 mark? Was that the original number polled in the experiment you described?
Thanks so much for the encouraging responses guys 
.
Sounds like a a great experiment to me. One question, how did you arrive at the 1,500 mark? Was that the original number polled in the experiment you described?
Thanks allot wolfe. No, the number of participants in the cattle experiment was by all accounts around 800.*
In deciding the number that would comprise this first organic network I was forced to find a number that we could realistically obtain in a decent amount of time but would also give us a good sized network. 1500 seemed about right to me.
Pava asked via Thepoint whether we had to wait till we had 1500 participants in order to get started. The answer is yes, though I realize that this may be bit of a downer to some of you. Think of it this way though: if we don't have 1500 people that are going to participate actively then we don't have a network or a concept to test. Besides, using a network of say, 25 people is not enough to cause the critical mass associated with collective intelligence.
We can get to 1500 pretty quickly though, given the speed by which information travels online. We need to network this idea around is all. I think we might all have at least one like minded friend, who might then have one like minded friend, and so on. Remember, if this works, we all get paid. No one person is going to be benefiting more than any other here.
In addition if any of you guys has a website thepoint has a function whereby you can feature the campaign there.
*The experiment is best recounted in the book The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations by James Surowiecki.
cool idea, there were some people trying to do this last spring, I just can't find the link for it right now....
cool idea, there were some people trying to do this last spring, I just can't find the link for it right now....
Thanks!
Hmmm...I must have missed that. It would be interesting to see how far they got. This is why I chose to launch via The Point, so there would be a public record of how many people joined the project.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Hey guys..Ive started a campaign at thepoint.com that I think you might have some interest in. The idea is to get a large number of people to participate in a series of experiments in which we will determine whether the concept of an organic neural network has any merit. Let me explain.
In 1906 or thereabouts a scientist by the name of Franci Galton oversaw an event in which citizens could pay to submit their guess of the weight of an Oxe for a prize for the one who got the closest weight. What Galton noticed was that while each individual person was usually way off the average of all of the individual guesses was less than .01% off from the actual weight of the Oxe. The collective performed far better than any individual ever could. This is one example of the phenomenon called "The wisdom of crowds". What I am proposing is simple:
We collect enough people to poll them on where they think the price of the EUR/USD will go in a few hours and I will take the trades on a paper account. The first experiment, which I am calling FOREX:Phase 1 requires 1500 participants to begin.
The great thing about this as I see it is that everyone will have varying levels of familiarity with FOREX and some participants will have none and yet we may still be able to create a great price prediction system. members of the Phase 1 experiment will be asked to simply look at a naked EUR/USD 1 hour chart (I'm recommending MetaTrader for this) and to input their first gut reaction to the following question: Will the price rise at least 25% in the next 5 hours, stay within the same range, or fall by at least 25% in the next five hours?"
We are attempting to collect their subconscious assessment of the situation. The theory is that the when the group has an overwhelming collective reaction then this will be a trade worth taking. We may then use technical analysis to locate the reasoning behind their collective belief. What we want to know is whether the collective subconscious can cues in the price line that we alone can not. I will be outlining the exact experiment protocol and the system's trading rules in the blog and I welcome any insights that you may have on how to go about this as I post my ideas. This is a collaborative effort.
Check it:
If we can demonstrate that this is a viable way to trade then we will have discovered a trading system that is completely free to all. You see, it wouldn't be a marketable system. No one person could claim a monopoly on a system based on collective intelligence. All you'd need to do to reap the benefits is to be a member of a big group of like minded 'nodes'. Each 'node' in the 'network' gives their input and individual 'nodes' in said 'network' could act on the group consensus if they so chose* because the collective input would be available to all equally. Thepoint.com is great for organizing this because it allowed me to post this proposal as a campaign. This way no one takes any action unless we get 1500 participants but once we get those 1500 we have our neural network.
Once we get going I will be making a new thread here for each trade that contains a poll. members of our organic neural network simply come here and answer the poll. As we gain steam I'm sure that we will implement more sophisticated ways of gathering data and analyzing it but I think to gauge whether there is anything to this hypothesis polls should be okay for a start. I will be posting the result of each trade and the eventual statement on the project blog. I'm thinking of actually taking the trades on a micro account but IDK.I'll let you guys know.
I mentioned that we can't get started until we get 1500 people interesed enough to join the campaign so let me make it 100% clear here that when I say "join" all I mean is that they (and you, hopefully) go to the URL below and click the "join" button. There is no cost, obviously, but I just want to be 100% clear on that. If you think this is a good idea then why not go ahead and tell a friend about it? We can get to 1500 pretty quickly with minimal effort.
These are the two questions that we are hoping to answer:
1. Can the "organic neural network" give profitable trading signals?
2. If not, can it improve it's performance over time?
Please see the blog for more information:
Applied Collective Intellegence
If you choose to join the Phase I experiment please note this is where i will be updating to.
You can join the effort here:
Collective Intelligance applied to finanical markets | The Point
*All participants act on the collective consensus at their own risk. I am not implying that this "organic neural network" will provide profitable trading signals. I am merely suggesting that we find out if it can. Please use your best judgment here. If you choose to make trades based on the collective consensus of the Phase 1 group please do so on a demo account.