The future of automated trading: round two - page 22
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
Better yet, OnCalendarEvent for handling calendar events;)
The discussion in this thread is high level - global perspectives on autotrading without reference to any specific software solution. It considers the antagonism of robots and small traders, and the changing nature of the market. The discussion of all other issues, and especially the discussion of what is and is not in MT, is in the garden, all in the garden, and you can measure bananas there.
Moreover, even if we go along with the trader, has anyone calculated the effectiveness of using tick history opportunities for the trader? That is, by how much more will the trader earn, using the tick history opportunity? Correspondingly, how much more efficient are the investments in this opportunity? )))
Whoever has all this information and processes it. 60 gigabytes a day in packed form.
This brings us to another question. How does an average trader get these 60 GB (well, let's not 60, but say 6, or even 3)?
Another question is whether brokerage companies (brokers) need such a load on the network, considering that at least 100 people can request this date at the same time (while the total number of clients may be in thousands).
It's not about the tick, as I understand it, it's about the existence of holes in the minute history.
No one is going to destroy anyone. Just trying to profit from a strategy by many participants reduces the profitability of the strategy.
A simple example, the same arbitrage: sell a stock and buy a futures on it if prices diverge by 2 sigmas - they will converge and there will be a good profit. But if many people want to buy at 2 sigmas it is clear that all will not be enough and the cunning one decides to buy at difference of only 1.5 sigmas, profit is smaller but guaranteed as it is the first. Now everyone who waited for 2 sigmas has nothing, and someone decides that profit from the divergence of 1 sigma will be enough for him. And so it goes all the way down to the minimum barely covering transaction costs, when only the fastest and the "unhungriest" can use the strategy, and that is a robot in a large company. And all individuals go to hell... And so with any strategy.
On the fund it will most likely be so, but speculators will be left (in my opinion) with Forex, moreover it is considered by most professional traders as purely speculative.
Well, let us not argue that private traders for the most part are investors and not speculators?
The topic of this thread is high - the global perspective of automated trading without reference to any particular software solutions. The antagonism of robots and small traders and the changing nature of the market are being considered. With all other questions, and especially discussions of what is and what is not in MT - in the garden, all in the garden, you can go bananas over there.
Here is this statistic for 10000 bars CO>3*spred ()=34 HL>4*spred ()=24 those in 34 bars the body of which exceeds the spread by 3 times Shadows exceeds the spread by 4 times only 24, and in 10 exceeds 3 spreads but no more than 4 and so on 5 spreads exceeds only 13 HL minute bars. 10 spreads no spread at all.
And what is 10 spreads 30 pips, those spreads exceeding 30 pips no spread at all, a 15 pips spread appears 13 times per 10,000 bars.
369 bars has a margin of error of 6 pips if measured on an M1 opener.
Nikolay, but you can't do it that way. It's not statistics but profanity. Your conclusions are that there are no minute bars more than sixty points (I'm not quite sure about the shadows; is it thirty points?). The sampling is for seven days only(!), you don't consider the weight of elements at all and don't consider real trading conditions.
Why bring the discussion back to the level of fifth graders?
But the general idea is clear - you can trade on the daily market without any stops and ticks. And there is no auto-trading at all. But we are not talking about the daily chart at all.
On so-called "market efficiency" from the perspective of real economics.
The theme raised is rooted in the theory of market efficiency. From the point of view of economics, rather than the mathematical theory of the free flow of information, this theory rests on the basic operating principle of the economic model of"pure capitalism". This is the principle of perfect competition. There are two necessary conditions for perfect competition to exist: first, there must be a large number of market players, and second, all market players must be approximately equal. Transferring these statements to the context of auto-trading, we have as participants a mass of trading robots trading on an equal footing, both in terms of their financial security and in terms of their access to information. No single market participant can create an edge in their favour because, firstly, they constitute only a tiny fraction of the competing forces and, secondly, they have access to information that is available to all. This is a theoretical model, very simplified, with no state regulation and with stock-exchange price quotes that are purely random. But in reality there is not a single state with a pure capitalist model. The USA is the most developed capitalist superpower but has a mixed model of capitalism. The US pursues an active monetary and fiscal policy, is a global member of inter-economic relations, and actively redistributes income among the population.
One way or another, the model of pure capitalism, is already heavily distorted by state action. But I believe that the basic distortion of the theory of pure capitalism and hence of the theory of efficient markets arises precisely because another basic principle of these models is not observed, namely that of the large number of participants in transactions and their relative equality. Based on this postulate, monopolies, price fixing and market price manipulation simply cannot exist in a world of perfect competition. This is because there are too many market players and each of them produces only a small part of the output, so there is no possibility for one or even a small group of such players to influence the actions of market forces due to the relatively small volume of their output. In the real world, different processes take place. The participants in market competition have a clear tendency to consolidate and merge. In the world of an efficient market, any serious consolidation would not be possible, because there are not enough resources to do this. Where would they come from, if there are no excess profits (the market is efficient and does not generate excess profits). Profits go to provide the costs (workers' wages, rent, interest, entrepreneurial fees, etc.). Indeed, in the real world most firms never grow to the size of large international corporations, but not many companies manage for some reason to overcome this barrier and become major international corporations. Either way, the modern, industrialised world as we know it has the hallmarks of capitalism, in which the vast majority of players are small, individual private firms in their organisational and legal form, and only a small proportion of the market players are corporations. But, and this is most interesting, the vast majority of the products produced (and hence the resources consumed) belong to corporations. Thus, according to economics (I always have this book handy), individual private firms account for 74% of the total number of participants, while corporations account for just 14% (the remaining 8% is taken up by the association of small firms, partnerships). However, the sales share of firms is only 6%, while the sales share of corporations is 90%! There is a significant skew from "perfect competition between many market players" to the global dictatorship of a handful of super-companies. The multitude of small firms, with their 6% share, have little or no influence on the action of market forces. And yet the theory of efficient markets places the main function of market regulator on them. In reality, the main regulators are a handful of global corporations which account for 90% of all goods produced and resources consumed.
For the reasons stated above, random walk theorists will never be able to prove the randomness of exchange prices, much less convince reasonable people that our industrialized world is a model of pure capitalism. The reality we are witnessing shows us otherwise. I can look at the amount of resources consumed and those who consume these resources at any time to see if I am right. For example, I looked at the data on gold for September 28, 2010. It is now in price and the eyes of thousands of traders are on it. Who holds the vast majority of gold contracts? Numerous market traders? As if that were not the case. 50.3% of all short gold contracts are held by only 8 market participants! These are not made-up numbers, these are weekly facts which are published by the USA. Think about this fact: half of all gold contracts, is in the hands of only 8 market participants! After that, can we say that everyone is equal among themselves, that everyone knows everything, the market is efficient, there are a large number of participants who individually can not affect the market, etc., etc.
On so-called "market efficiency" from the perspective of the real economy.
About so-called "market efficiency" from the perspective of the real economy.
Don't confuse the warm and the soft. And most importantly, don't argue against your own ideas. Firstly, no one has said that market participants must be absolutely equal. The implication is that they are equal in access to information, but not in the amount of funds. The classics naively assumed all investors to be rational, which is not the case and which only makes things worse. Secondly, no one has ever mentioned strong market efficiency. Today's markets exhibit a weak degree of efficiency. And even the increasing efficiency of markets, which can be objectively observed, leaves the market in the same phase of weak efficiency.
And the increasing efficiency is also due to robots from big companies. So for large companies the markets will remain inefficient, just as arbitrage still exists for them. But for everyone else, the markets will be efficient: no arbitrage, no TA, nothing systematic - just roulette.
Well, it all kind of adds up:
1) we make mistakes in simple maths (because in reality we never counted, banal erudition and confidence were enough), we call the mistake 10 times "forgot the zero" (such non-recognition is symptomatic).
2) stubbornly unwilling to count real volumes (years of history, tens and hundreds of instruments, tens of thousands of users), stopping at "well, look, not so much!", forgetting to clearly spell out"don't believe me, I did the calculation in 24 hours".
And yet they naively try to mislead the rest of us. And this is the main mode of defending their position: to underestimate and compare with incomparable (amazing examples of comparisons with wild distortions are present on this forum - one "make decisions once an hour, rule once an hour" stand out). Economic realities are generally left out, it's easier to live in an invented world.
Is that all you're saying? Maybe you missed this post https://www.mql5.com/ru/forum/2176/page21/#comment_26385 it's interesting, I'll answer for zero.
To point 1. (forgot the zero), it wasn't me who already replied. but I'll cite it here perhaps you missed it.
LeoV:
Wrong - 60*60*24=86400 ticks per day. Respectively, 86400*36=3110400.
gip 2010.10.01 23:08 Terrible. Now compare this data flow, for example, with the flow of some Skype. They also say that there is Internet television. They must be lying.
On point 2. you are absolutely right. I didn't write "don't believe me, I did the calculation in 24 hours". Correct "don't believe anyone". Check it's the volume in a day that needs to be transmitted, technically not a feasible task. Gip is right - internet television does not exist, because this volume of data has to be transferred in 1 second, not in 24 hours, but in 1 second.
There's a thread about the future, here's the future http://www.cisco.com/web/RU/news/releases/txt/2010/060310.html.
On point 3. About realities. I tried to explain complicated mathematical constructions in simple and clear language, with examples that are accessible and understandable to most of the forum participants. I would be happy to communicate in the language of formulas, but unfortunately the level of my opponents does not allow. Example - please, what is the difference between the Pontryagin's optimal control criterion (OC) and the Letov-Kalman criterion, which of them is more applicable to the finnish markets?
So I have to explain it with examples. The phrase you took out, "Decide once an hour, rule once an hour" is worth a lot.
I can explain it in MQL.
Many EAs have such a check and if the EA, for example, works on hourly candlesticks, then this code exactly corresponds to this phrase. This is absolutely wrong from the MA point of view and common sense. The realities as you put it are different, the market can turn at any time, not at the end of the hour. The market has already turned and we are smoking, waiting to touch the wheel (get out of the deal or flip).
Z.I. DO NOT BELIEVE ANYONE. check. Prival the bastard lies, says on ticks better, work on candles = touch the steering wheel 1 time per hour when driving on a mountain road and happiness awaits you...around every bend....
Once again you are trying to get away from the heart of the calculation.
We are talking about deep tick history, not tick flow per day. And the tick history is exactly the volumes I have shown. Gigabytes per user at any request of an army of thousands of traders.
You are too far removed from practice, technology and economic feasibility. That's why you earned the "don't believe me" label to your posts a long time ago.