The future of automated trading: round two - page 17

 
rsi:
That we don't fight, Leonid. timbo defends the hypothesis that an efficient market does not allow predicting the movement of any instrument based on past information. Therefore, the poor trader has nothing left in the future life - all will be grabbed by the giants. So I wonder, how do your networks work? :-)

Read the Russian description, it's enough for me...

The market efficiency hypothesis (GER):

3. Strong form: current asset prices take into account all information from both public and non-public sources and also take into account non-public (insider) information, such as that available to the managers of a company regarding the company's prospects. A strong form includes both a weak and a medium form. A market that is effective in its strongest form can be called perfect - it is assumed that all information is publicly available, free of charge, and reaches all investors at the same time. In such a market, it is meaningless to make investment decisions even on the basis of insider information.


1. As far as I understand from this, the strongest form includes the other two, and it is explicitly stated. Or am I wrong?

2. As far as I understand, some history is available for all THREE forms of performance. Another thing is public or insider information (these performance indicators may not be available to a certain number of participants). Or am I wrong?

3. A COMPLETE and therefore strong market is purely a theoretical thing. Because there are no IDEAL conditions for the dissemination of information and the preparedness of the participants.

Especially it concerns "closed" (insider) information. And publicly available information sometimes can not get at the right time, or it can not be considered objective.

Or am I wrong?

4. As far as I understand it, the use of mechanization, on the contrary, reduces the effectiveness of the market - because a bunch of disparate machines trade without paying attention to news, insider information and other fundamentals. This theories the entire essence of FA (because in its conventional form it is lost to the analysis of the situation).

At the same time mechanical systems downplay the role of technical analysis itself.

PS

The last point personally leaves me with a lot of questions.

 

sandex: Проблема в том, что не за что покупать. Да и не к чему. 

I agree. I'm not a fan of tick data either. But if someone wants one, what's the problem with buying one and torturing them all they want? Metaquotes don't want them - that's their right, not their obligation ))))
 
timbo: Well, he has already answered this question - they work sluggishly, there are no super-profits, as well as happiness in sophisticated networks.

It's not about super profits. 100-200% per annum is good business, which you can easily get from Forex. Some people want 1000% a month. Well, as they say - there is no harm in wanting))))

 

By the way, 1 day's worth of tick data will weigh about 100 meg. Ready to digest such volumes? )))

 
LeoV:

By the way, 1 day's worth of tick data will weigh about 100 meg. Ready to digest such volumes? )))

A full financial date - all trades in 150 markets from Reuters weighs 60 gigs a day. But someone is digesting it all. I don't think he has a problem with his stool...
 
timbo:
The full financial date - all transactions on 150 markets from Reuters weighs 60 gigs per day. But after all, someone is digesting it all. I don't think he has a problem with his stool...

No problem. In my opinion, a story of this quality is needed for mere mortals to a depth of no more than a week, for major players to a depth of a month.

Here are some preliminary calculations of the necessary disk space:

1. For mere mortals (although not very simple, given that the story is bought from Reuters) - 300 GB = 5 days * 60 GB (I have for example on a working machine 2.5 Ter)

2. For large players (in my opinion) - 1200 Gb = 20 days * 60 Gb.

Do you think I can not pull up on disk space?

PS

Another thing is how to load all this date?

Equally interesting question is what kind of deposit (income) should someone who uses Reuters services have...

 
papaklass:

Gentlemen, do you take into account the fact that there is fierce competition between the giants of the industry. And if someone comes out ahead, they are slowed down by any means necessary. The annual accounts of these giants prove this point. And when the giants fight with each other, then we, simple traders, will always find a place in the markets, where we can get a small piece for our living. Timbo, so not everything is as hopeless as you imagine.

The problem is that in order to get something out of the market you have to get there and you have to do it so that a beginner doesn't get swept out of the market in the first month ...
 
papaklass:

Take care of the losses, otherwise they will take care of you. This tenet has never been undone.

It's not about the losses, but about the opportunities and the strategy a newbie can implement in the market. Somebody enters the market with $1000, somebody needs $10,000, and somebody has $100,000 to start with.
 
timbo: All trades in 150 markets from Reuters weighs 60 gigs for the day.

Who said that? I just downloaded Eurodollar ticks from eSignal for the sake of experiment. Three days - 150 meg.

Files:
01oct.rar  1165 kb
 

And this is today's day, as it's not over yet, weighs 24 megs. But yesterday's full day already weighs 53 megs. So try to process this and then write....)))

Files:
30sept.rar  2481 kb
Reason: