Where to start? - page 17
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
That wasn't my intention... on the contrary, I put a smiley at the end of my previous post... If the humour was unfortunate, please excuse me...
I still don't get it, does the top starter have any basic knowledge in the areas:
1) trading;
2) programming (?)
"basic dc"
"has the intention, started reading the literature"
And let's discuss the factors that can objectively reduce a beginner's efforts "to become a profitable trader" (algotrader) to naught.
It is a matter of habit, if they had started counting on their feet 10,000 years ago, it would be convenient now, but otherwise .... that's the practical point of having 10 toes... a few toes would be like hooves, less toe sweat :-)
It's clear here. The stick has two ends, and everything in between is politics, philosophy or just "can we talk". The dualism of the universe lies in two highlighted absolute opposites. Life-death are two absolute opposites, like one and zero. Half-living, half-dead, slightly alive, etc. - this is in the realm of medicine, religion or politics, not mathematics. Light and darkness is the absolute opposite, dusk is neither fish nor meat, philosophers will long argue that this is a "child of light" or a "child of darkness". God and devil are absolute opposites, man as God may create and as devil destroy all, in general neither fish nor meat.
Two ends without middle are two sticks.
The state of clinical death can reach several hours, with equal probability to survive or die. The morgue? And embryonic life? Since when is alive - abortion is forbidden - and an hour before was still dead?
Dusk is the state in between. Morning and evening - one third of the time of day.
There is no room in the soul of an atheist for God or the devil.
Neutron - proton - electron, unstable equilibrium state, etc....
Dualism describes an extreme state, one. Triangles rule.)
PS It's a matter of personal outlook, of course.
And let's discuss the factors that can objectively reduce the beginner's efforts "to become a profitable trader" (algotrader) to naught.
That wasn't my intention... on the contrary, I put a smiley at the end of my previous post... If the humour was unfortunate, please excuse me...
I still don't get it, does the topicstarter have any basic knowledge in the areas of:
1) trading;
2) programming (?)
"basic dc"
"has the intention, started reading the literature"
Quite right! You cannot take it away from them.
Discussion of pitfalls, this would be of interest to all newbies, not just me I suppose.
Inability to learn, lack of perseverance, negative impression of the first contact with the technical support team, opinion of others, laziness .
The first 2 and 4 are not available, the 3rd hasn't been tested(what's so scary???), the 5th is inversely proportional to interest, at the moment interest is off the charts and laziness is gone, I think it's for the long haul.
In one of the neighboring topics I asked about the nuances of strategy testing, but no one gave me any detailed answer((( I think this is also a great obstacle. Because if the testing does not correspond to real trading, what is there to rely on?
In one of the neighbouring threads I asked about the nuances of strategy testing, but no one has answered anything substantively((( I think this is also a big obstacle. Because if testing does not correspond to real trading, what should we rely on?
It all depends on the purpose or why or what they are looking for or what they are trying to find by testing..... many options...
For example 2 cardinal directions.
1 Testing of the Expert Advisor for the purpose of studying the Trading System
2 Testing of the Expert Advisor for the purpose of optimizing the Expert Advisor code
...Again meaning manual testing of a trading strategy or software....
It all depends on the purpose or why or what they are looking for or what they are trying to find by testing..... many options...
For example 2 cardinal directions.
1 Testing of the Expert Advisor for the purpose of studying the Trading System
2 Testing of the Expert Advisor for the purpose of optimizing the Expert Advisor code
...Again we are talking about manual testing of a trading strategy or software....
I'm not really equipped to have that kind of conversation right now. Maybe I meant automatic testing in the mt5 tester.
I just wondered how the testing of a very simple Expert Advisor in 2 modes can be so different. OHLC and ticks.
Like heaven and earth. The question arises as to why? Which one is most similar to the real one? I don't know why I need OHLC, if it is inadequate.
I was assuming that testing is a summation of the profits of all trades in a period minus the costs. And it should have a link to real trading. Otherwise, what is the point of it?
It would be logical to adjust the tester in such a way that there would be minimal difference from the real trade, but how? What is not taken into account?
I'm not very well prepared to have this kind of conversation right now. I probably meant automatic testing in the mt5 tester.
I was just wondering how testing a very simple Expert Advisor in 2 modes can be so different. OHLC and ticks.
Like heaven and earth. The question arises as to why? Which one is most similar to the real one? I don't know why I need OHLC, if it is inadequate.
I was assuming that testing is a summation of the profits of all trades in a period minus the costs. And it should have a link to real trading. Otherwise, what is the point of it?
It would be logical to adjust the tester in such a way that there would be minimal difference from the real trade, but how? What is not taken into account?
For about a week now, I have occasionally wondered "how to estimate the time and monetary costs of training ... ". Nothing popped into my head, but perhaps now I had a thought in the right direction.
You say "turtles", but I say the teachers there were driven by fame, when you have money you want to be recognized. I do not think so here, because more and more people feed the elks, and therefore the training for a quid can be forgotten. You can forget about time spent on education, N years of troubleshooting and searching for something that exactly suits you, you can't touch that for a beginner. And so, if the fame is minus, the formula "time-money" time is also not returned, there are lots of losers. Let's say someone has caught a lot of moose in N years, 3 grand quid, now let's say since the beginning of the year he's been in the plus. I think it's fair to evaluate his knowledge into all the provable moose on real accounts. The plus side of this formula is an individual approach in my opinion and some kind of moral compensation for the teacher at that time, while he will transfer the knowledge, and the students will save a lot of nerves, perhaps, however, it is up to them to decide. Lastly, this idea goes well with Livermore's words that there is nothing better in the market than to lose a deposit, it teaches well not to do what is not necessary :)
Moose is counted using the formula "Input - Deposit at the beginning of the year" for each account. Let's say T(teacher) deposited a grand 3 times, at the beginning of the year he had 1000, now he has more than 1000, say 2000. The tuition costs 2,000.
Mind you, I'm not claiming anything, I'm not the truth, I'm not saying there can't be other formulas, I'm not saying you can't learn differently, I'm not saying you can't fiddle with counting real accounts, I'm not saying the knowledge imparted is all the pupil needs. My point is simple - to give as simple a calculation as possible of the amount of tuition on the part of the teacher, for what money he can share the experience. There is no guarantee that for that money you will get the knowledge that suits you, just as there is no guarantee, by the way, that you will be able to rip off that market crowd
If anyone feels like it, you can think it was a throw-in. Maybe it really was.)
See you at the market!
Is there no answer?
Thank you. The picture is partially complete. I read about generation of ticks at https://www.metatrader5.com/ru/terminal/help/algotrading/testing. It turns out that due to limitations of minute quantization, one can only start from minutes, the results inside minutes are meaningless.
Frankly speaking, I do not understand what for we need such an algorithm of tick generation for OHLC candlesticks. If there is only OHLC data for one minute, it makes sense to consider the arithmetic mean of OHLC as the most probable value that can be registered there for sell and +spreads for buy. I don't understand the deeper meaning of this:
No one knows how it's wobbling inside the candle if there's no data. Why should we complicate things. All the more so that the structure is rigidly defined. I must have misunderstood something...
But if the algorithm somehow uses the value of a nonclosed candlestick generated by the algorithm of tick generation, it follows that it will try to interact with the constant structure of price behavior which doesn't exist in reality. So, ticks are something else after all. When reading the manual of the terminal I did not think about it.
Ticks are absolutely different in different brokerage companies. In some cases they are even artificially created. You can see it from the tick chart.
Thank you. The picture is partially complete. I read about generation of ticks at https://www.metatrader5.com/ru/terminal/help/algotrading/testing. It turns out that due to limitations of minute quantization, we can only start from minutes, the results inside minutes are meaningless.
I also find such artificial generation of ticks redundant. If there is a good trading strategy based on reasonable timeframes (not M1) the minute OHLC for testing will be enough.
Theoretically it is very easy to test the strategy for robustness or rather how much the lack of information about real intra-minute tick distributions will affect the trading system.
We know for sure that the Open price appears first and the Close price last. We do not know only in what order High and Low were formed.
Therefore we can test the system in two modes: in the first mode High is formed first, then Low; in the second mode it is vice versa. That is all! No matter how the price moves inside the bar, whatever the patterns are, we will anyway see either the first or the second variant. That is quite enough. If the results are approximately equal, it doesn't matter
what was the real tick distribution in a minute candle. It's just market noise.