Artificial Intelligence - page 2

New comment
 
Shahab Shamsi #:

hi friend!

I'm not agree with you! we have a normal distribution in mathematics which is made by random data. however it's not completely Gaussian but you can claim it's a semi random data. what you say is the patterns which made by predictable data section.

Dear Friend!

semi random market nature is proved by econometric models. it's not there were any hesitation about. it's like I explain you the earth is not the  center of the world. any way you can make a random walk model and make your own candles by it's data. you will be surprised when you see you may have even trendlines! however I don't claim all data in market is random. you should try to find it's predictable and useful data and give out the useless data in your analysis. it's like a book which has many blank pages on it. you should try to find the pages which have information on them.

 
Shahab Shamsi #:

Dear Friend!

semi random market nature is proved by econometric models.

Any model is based on a premise. What happens though when the model is based on a false premise? This is a philosophical question which cannot be easily answered. If you want to subscribe to the belief that (semi)randomness exists, then that's fine. However, my experience both within and outside of the financial markets, has shown to me beyond doubt, that randomness is an illusion which we use as a descriptive term to classify something that we perceive as chaotic. Now, what if chaos is a higher form of order? Wouldn't decrypting and perceiving that order disprove the very notion of chaos? In other words: wouldn't that mean that there is an underlying pattern to what we call "random"? 

Shahab Shamsi #:

it's like I explain you the earth is not the  center of the world. any way you can make a random walk model and make your own candles by it's data. you will be surprised when you see you may have even trendlines!

The random walk theory is demonstrably false. Markets are cyclical and fractal in nature, and there is ample evidence for this. It is not a coincidence that academics in their ivory towers often come up with theories that are easily debunked by practical reality. Please excuse me if that sounded cynical, but I have no high opinion of these armchair-theorists who have absolutely no industry experience but are always quick to tell others how things "really" work. If the random walk theory was correct, then we wouldn't have any investors making near-consistent returns in the markets. People like Warren Buffet would be flipping a coin before deciding to invest. Hedge funds trading the FX markets would be flipping a coin. Actually, nobody would even put a single dollar into these markets if they were random and have better chances going into a casino and try their luck there.

Shahab Shamsi #:

you should try to find it's predictable and useful data and give out the useless data in your analysis. it's like a book which has many blank pages on it. you should try to find the pages which have information on them.

Exactly! It's called pattern finding. You fetch enough useful data, formulate a prototypal trading system based on that data, run that through an algorithm that trades that pattern with both historical ticks, as well as Out-Of-Sample ticks and if you end up with more profitable trades than losing trades, you have a potentially workable, profitable trading system. That's the modus operandi of HFs and IBs.

 
Suren Khosravi #:

Any model is based on a premise. What happens though when the model is based on a false premise? This is a philosophical question which cannot be easily answered. If you want to subscribe to the belief that (semi)randomness exists, then that's fine. However, my experience both within and outside of the financial markets, has shown to me beyond doubt, that randomness is an illusion which we use as a descriptive term to classify something that we perceive as chaotic. Now, what if chaos is a higher form of order? Wouldn't decrypting and perceiving that order disprove the very notion of chaos? In other words: wouldn't that mean that there is an underlying pattern to what we call "random"? 

The random walk theory is demonstrably false. Markets are cyclical and fractal in nature, and there is ample evidence for this. It is not a coincidence that academics in their ivory towers often come up with theories that are easily debunked by practical reality. Please excuse me if that sounded cynical, but I have no high opinion of these armchair-theorists who have absolutely no industry experience but are always quick to tell others how things "really" work. If the random walk theory was correct, then we wouldn't have any investors making near-consistent returns in the markets. People like Warren Buffet would be flipping a coin before deciding to invest. Hedge funds trading the FX markets would be flipping a coin. Actually, nobody would even put a single dollar into these markets if they were random and have better chances going into a casino and try their luck there.

Exactly! It's called pattern finding. You fetch enough useful data, formulate a prototypal trading system based on that data, run that through an algorithm that trades that pattern with both historical ticks, as well as Out-Of-Sample ticks and if you end up with more profitable trades than losing trades, you have a potentially workable, profitable trading system. That's the modus operandi of HFs and IBs.

Dear Friend! I agree with you that you can get constant profit in the market. but it doesn't mean that market is completely predictable! you may have an image which is noisy. but you may still detect the image behind it. what is important in this market is that not all data has the same value. you should detect meaningful data from your market data and try to analyze according to it. this is really what technical method do! when you draw support and resistance or trendlines or fractals you accept this priorities! in fact you try to analyze market according to it's critical points. in Neural Networking  we do the same but with mathematical method. however it's a very important procedure which can affect completely on our results!

 
Suren Khosravi #:

Any model is based on a premise. What happens though when the model is based on a false premise? This is a philosophical question which cannot be easily answered. If you want to subscribe to the belief that (semi)randomness exists, then that's fine. However, my experience both within and outside of the financial markets, has shown to me beyond doubt, that randomness is an illusion which we use as a descriptive term to classify something that we perceive as chaotic. Now, what if chaos is a higher form of order? Wouldn't decrypting and perceiving that order disprove the very notion of chaos? In other words: wouldn't that mean that there is an underlying pattern to what we call "random"? 

The random walk theory is demonstrably false. Markets are cyclical and fractal in nature, and there is ample evidence for this. It is not a coincidence that academics in their ivory towers often come up with theories that are easily debunked by practical reality. Please excuse me if that sounded cynical, but I have no high opinion of these armchair-theorists who have absolutely no industry experience but are always quick to tell others how things "really" work. If the random walk theory was correct, then we wouldn't have any investors making near-consistent returns in the markets. People like Warren Buffet would be flipping a coin before deciding to invest. Hedge funds trading the FX markets would be flipping a coin. Actually, nobody would even put a single dollar into these markets if they were random and have better chances going into a casino and try their luck there.

Exactly! It's called pattern finding. You fetch enough useful data, formulate a prototypal trading system based on that data, run that through an algorithm that trades that pattern with both historical ticks, as well as Out-Of-Sample ticks and if you end up with more profitable trades than losing trades, you have a potentially workable, profitable trading system. That's the modus operandi of HFs and IBs.

sometimes we may try to judge scientific theory according to our philosophical thoughts. it's completely wrong. philosophy and science are completely different. when your data is very large and it's distribution is Gaussian you can't say I think it's still meaningful. without esteem to the science you can't expect your progress!

 

Hi Shahab Shamsi,

I think the machine only as a tool to give your information, the executor is ourselves, we decide either entry or close position on the market.

What do you think Shahab?

 
Hansen #:

Hi Shahab Shamsi,

I think the machine only as a tool to give your information, the executor is ourselves, we decide either entry or close position on the market.

What do you think Shahab?

it's depend to your system. there is no limitation for AI bots. they can analyze and decide to trade better than human. they don't have emotions and can analyze more powerful. so I think robots are more trustful.
 

Forum on trading, automated trading systems and testing trading strategies

Taking Neural Networks to the next level

Sergey Golubev, 2024.03.07 13:47

Neural networks made easy (Part 60): Online Decision Transformer (ODT)

The last two articles were devoted to the Decision Transformer method, which models action sequences in the context of an autoregressive model of desired rewards. As you might remember, according to the results of practical tests of two articles, the beginning of the testing period saw a fairly good increase in the profitability of the trained model results. Further on, the performance of the model decreases and a number of unprofitable transactions are observed, which leads to losses. The amount of losses received may exceed previously received profits.

 

Forum on trading, automated trading systems and testing trading strategies

Better NN EA

Sergey Golubev, 2024.03.07 13:50

Neural networks made easy (Part 61): Optimism issue in offline reinforcement learning

Recently, offline reinforcement learning methods have become widespread, which promises many prospects in solving problems of varying complexity. However, one of the main problems that researchers face is the optimism that can arise while learning. The agent optimizes its strategy based on the data from the training set and gains confidence in its actions. But the training set is quite often not able to cover the entire variety of possible states and transitions of the environment. In a stochastic environment, such confidence turns out to be not entirely justified. In such cases, the agent's optimistic strategy may lead to increased risks and undesirable consequences.

12
New comment