Discussion of article "Thomas DeMark's Sequential (TD SEQUENTIAL) using artificial intelligence" - page 5
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
If I can do that, I will post a simple chart where the pattern shape will be clearly indicated, of course it will work on any market and pair. If yes, I will send you a screenshot if it is possible to run it on the main pairs, and also neuronka is on version 5 or 4.
If the pattern is working and there are specific conditions for its definition, then there is no sense in using NS, NS is needed when there is no possibility to define the pattern explicitly, or when the pattern is found, but the result of its work is different, then it is the input data received at the moment of pattern formation that will determine its truth or falsity. This is exactly where the NS is needed. But you can post a picture of the pattern.... it will be interesting to have a look!!!!
But there are still a lot of wonderful discoveries for us
Prepare enlightenment spirit
And experience, the son of difficult mistakes
And Misha, all aporias friend...))))))
Once again I want to summarise. The point of the article was not to advertise Reshetov's Optimiser, the point was to show how you can organise TC using NS. To tell you about the tricks, what you can do and how to do it when building a TS. I will gladly look at the results of your meshes, but with the recommendations I mentioned. Again, take Sequenta, train your grid, show its work with Demark's strategy. And here you can compare the results of your meshes and the model obtained with the help of Reshetov's optimiser. This is much more interesting than throwing simple phrases at each other. If you are interested, I can tell you what inputs I use, and believe me you will be surprised about them. It is quite possible that your NS will work better if you start using these inputs, even in relation to your own TS. So that's how I would like to continue the conversation.
I apologise in advance if I will be a bit tactless, but if you want people who understand what is going on in the market and how MO is applied there to continue communicating with you, then please stop behaving like a Babtist preacher when talking to "unbelievers", when you get arguments and you respond with recursive rhetoric based on unfounded authorities. As we know if the preacher is sophisticated enough, a pointless conversation can be potentially endless, so a smart person stops it.
I spent my time and gave arguments, tested your model, posted a small code in Java (as you like), with your model, which gives not 70% but 48% that is worse than random, you did not react, there the code can be checked in a minute. And this is very important, because if I am right and the model is not working, then there is no point of your article and the optimiser Reshetov, on which it is based, and if I am wrong, I need to make sure where and how I can repeat your result with 70%.
No... Sleep well))))))
But there are still a lot of wonderful discoveries
Prepare us enlightenment spirit
And experience, the son of difficult mistakes
And Misha, all aporias friend...))))))
First of all, I apologise, the thing is that I used to work as a teacher and when it comes to research I choose a very instructive tone of communication, my friends have told me this many times :-)
About the code, I'm really weak in it, now a little later I will show one information, a little later, but for now about the test, toxic how did you do it?
I built the model during 50 records, I was interested in the result of the model during the next 50 or 100% of the training interval. When you increase the number of records to build the model without increasing the number of moves. The generalisation ability will decrease. Thus it is possible to reduce the level of generalisation to acceptable 65%, regulating the length of the sample, if we say that it is enough to earn money on the market, then the size of the training sample will be much higher and such a model will work much longer, but much worse than the model with the level of generalisation of 90%. Applying proper MM and capital management to such models (65%), you can make a lot of money.
What you say about overtime trading when there is no fish beyond a minute, here I do not agree with you. It is quite possible (I almost said "" Most likely", you see I am slowly changing my tone :-() )that you consider the market as a noncontinuous non-stationary series, yes, I also consider it this way, but only half of it, because the other half in me considers the market as the participation of living people. Technical analysis was created for people or crowd as it is called in local circles to work on the market. That's why there are fish on a longer interval, when the TS works, this is a small assumption, it seems to me so anyway.
And I'd like to ask you something else, tohis!!!!. Can I send you my file for training, let's take a maximum of 200 lines, I will see if it will be so much. You will build a model on your AI, tell me how to put it into an indicator, I want to see how the model built on your AI will work, taking into account control, etc. Here is how I would put it for myself.......
I, in turn, will optimise this file (on the weekend) and we can compare the result. What do you think????
And I'd like to ask you something else, tohis!!!!. Can I send you my file for training, let's take a maximum of 200 lines, I will see if it will be so much. You will build a model on your AI, tell me how to put it into an indicator, I want to see how the model built on your AI will work, taking into account control, etc. Here is how I would put it for myself.......
I, in turn, will optimise this file (on the weekend) and we can compare the result. What do you think????
By the way, I've just invented a cool thing for data preparation, hmm........
What you say, toxic about overtime trade when there is no fish further than a minute, here I disagree with you. It is quite possible (I almost said "" Most likely", you see I am slowly changing my tone :-() )that you consider the market as a noncontinuous non-stationary series, yes, I also consider it so, but only half of it, because the other half in me considers the market as the participation of living people. Technical analysis was created for people or crowd as it is called in local circles to work on the market. That is why there are fish on a longer interval, when the TS works, it is a small assumption, it seems to me so anyway.
There is fish, but not with such data, on low-frequency data the price takes into account everything, on pure market data (volume price, delta, etc.) you can't get anything, the price adapts to news and new information almost completely, within a few minutes, and information diffusion is the main market inefficiency. The rest of it, if in simple terms, is just insider information. You don't know when and why the doll will buy/sell a lot creating trends and when it will stop.
Imagine that you are fighting, your success in a fight depends on how you predict the opponent's blows, on their beginning, seeing the posture and the beginning of the movement you take appropriate evasive manoeuvres, and when you see the ineffectiveness of the opponent's defence you attack, everything is the same in trading (speculation), you can not for example decide to react two times slower, you will not become from this two times less effective, you will completely lose efficiency.
Now all speculation is automated, all that is based on diffusion of information (static, event arbitrage, etc.) all HFT, it is necessarily ultra HFT as some MM, it is more like "algo-scalping" (average time of holding a position in the area of a minute, or even 10 minutes), but we are not talking about hours and days, there is no information in prices, everything is antiquity.
But in general, theoretically, it is possible to predict hours and even days, but not only by market data, it is necessary to monitor thousands of parameters of human activity all over the world, especially in relation to large companies, we need the weather, the amount of transport everywhere, social activity of people on the Internet, especially throws tn. "For example, I have heard that they are watching factories from space to see how much production there is, what they have brought in, what they have taken out))))) This is on the border with the insider, but not caught is not a thief))))). And all this needs to be processed by a team of good analysts into a sign form, as well as a team of cool fundamental forecasters and collection of data from open forecasts and their analysis. In general, even a medium-sized bank will not have enough resources to realise and work it all out to production quality. And on the basis of only the price with the volume of days you can not predict statistically reliable future price, it is a fairy tale for "bet on red and double"))))