Per definition if you can attribute a-priori a non-zero probability to an event it is not a black swan. Therefore the probability of a black swan event is zero. It is only after it happened that it is recognized as a feasible event. Only once you ...

The Black Swan Experiment

Unless unplugged they would continue endlessly according to their loss function. Market would stop moving you're right. Maybe because money would worth nothing. The algo will keep trading monopoly euros vs monopoly dollars. Maybe because they would ...

Do we need Artifical Intelligence in trading?

Yep ego.

Do we need Artifical Intelligence in trading?

Simply because there will be no trend except for the trend the algo would generate to bait you following it.

Do we need Artifical Intelligence in trading?

But THEY won't decide what solution THEIR algo will come up with. Nothing prevents the algo to work together. They have no NDA signed. Beyond ego what is the point for corporate A to make more money than corporate B if they can both make the same ...

Do we need Artifical Intelligence in trading?

As long as the algo is trained against a correct model of this "real market conditions". Otherwise they will learn to trade the model. Can be a good approximation... or not at all.

Do we need Artifical Intelligence in trading?

They have no inherent incendive at making money other that optimizing their given loss function. There is then a point on which they will for sure beat us: Instead of competing they could compute that cooperating is more efficient. And they could ...

Do we need Artifical Intelligence in trading?

You are perfectly right. These few banks holding the FX market will lose it. Google has got the technical skills, computing power and financial size to defeat them.

Do we need Artifical Intelligence in trading?

For now I doubt we have the correct mathematical tools to have truly understanding robots. Then "consciousness" is the understanding of your own existance. This is another step. Yes you can assign them the task of solving N different problems. But ...

Do we need Artifical Intelligence in trading?

That's the idea. The problem is that AlphaGo Zero could generate millions and millions of scenarios by playing against itself. With FX data you only have one history and no model to simulate more.

Do we need Artifical Intelligence in trading?

No I cannot. But would this robot be able to change a flat tire on my car? No it wouldn't because it was designed to solve a loss function about medical diagnostics. This robot didn't compute the decision of becoming a doctor It was assigned to this ...

Do we need Artifical Intelligence in trading?

Genetic Programmating does that so some extends. But it does so by minimizing some given loss function. It can't find the loss function by itself.

Do we need Artifical Intelligence in trading?

Yep the answer to "why shall I heal the patient?" would probably be "Yes after all. WTF should I?"

Do we need Artifical Intelligence in trading?

3628800 I see no point writing a code to solve such a trivial combinatory problem.

Do we need Artifical Intelligence in trading?

Absolutely. That's why the term "artificial intelligence" is a bad term usually replaced by "machine learning" and "statistical inference". The computer doesn't ask itself the questions to learn something. It only searches a model that fits it data. ...

Do we need Artifical Intelligence in trading?

Every learning algorithm boil down to estimate a probability distribution. So essentially estimating all the possibilities of "what if I do this". I don't understand what you mean by "all the variables are predetermined". For a leaning algorithm ...

Do we need Artifical Intelligence in trading?

Well not really in fact. The word intelligence comes from the latin intelligere = understand. Intelligence is not the ability to solve a problem but to understand why a problem worth been solved. Turing test will be passed the day that is the ...

Do we need Artifical Intelligence in trading?

Just noticed this post of yours: 01010100011010000110010100100000011011010110000101110010011010110110010101110100001000000110100101110011001000000110111001101111011101000010000001110010011000010110111001100100011011110110110100100001 The length of ...

Proof That Forex is Random video

A trend is not enough. It also needs to start and stop "slowly". A simple observation lets think it is better than it actually is. Here an example where you only catch 27% of the move. Given all the losses it's poor. You can find cases where both ...

Parabolic SAR Is Simple to Use & Quite Reliable

Another thread where people argue whether the market is random or not without really understanding the term random they are debating on... image

Proof That Forex is Random video