Our Masha! - page 12

New comment
 

Dear Grasn.

What you are writing about in 2009 is long gone in the mid 1970s.

All these screenshots and simulated noise are not relevant to the financial markets...

as there are different patterns and distributions than pseudo-random normal.

So this 30+ year gap will have to be caught up, and at the root of the approach is a snag.

I know so many traders overseas without a quantitative base, and believe me, they do just as well....

As for your research so far just can't see what it gets you on the financial wv. ranks....

If you want to make trading your hobby, then the conversation is useless as you are being spoken to by professionals who have been in the market for more than a year.

Propose your invention and run tests, if it really makes money, you will get recognition and money very soon.

Imho, when it comes to solving symbolic equations, such as SRS, ODE, etc., there is nothing better than Maple.

For dealing with time series, maths is better. For industrialization of calculations better and simpler than Matlab...

PS unfortunately, from my experience pilot + forex is hardly compatible....

 
Quant >> :

because there are different patterns and distributions than pseudo-random normals.

I think the grasn clearly understands where there are normal and non-normal distributions.

If you want to make trading your hobby, the conversation is useless as you are being spoken to by professionals who have been in the market for more than a year.

Excuse me, Quant, who are you talking about?


P.S. 2 BARS:

They take pleasure in screwing themselves and others brains out in higher mathematics.

What's up, Michael? If you don't know about "higher maths" - why do you call it "you"?

 

Well, that's understandable,

but why take a sword to a tank...

Mathemat,

about those who earn their bread working in this field.... proprietary traders...

 

to Quant.

In case you haven't noticed, I'll try to correct your visual defect - I've talked to Prival, he has a lot written in MathCAD, and Neutrona as well, and this tool will allow to integrate products perfectly. And all this crap about the seventies and so on - what's that about? Have you confused wheels, have you pressed the wrong pedal? There are excellent tools, very briefly described - and on you "long gone in the mid-1970s", "different schemes and distributions here" ... . Which ones are different? Where did you read about what I wrote? Have you even read the posts? Do you really think that I'm going to expose my methods and approaches to no one? What kind of bullshit is this Quant?


to Mathemat

Good to see you're awake too! :о)))


PS: ohhhh, I've started the theme again :o)

 

Apologies to the esteemed thread creator for the off-topic...

As for the instruments, of course I have nothing against them.

Further, about the 70s, there's a question of an approach to making money. putting a quote through a filter is certainly good, but it's weak...

Grasn, I liked your thread about the asset management system.

"I hope the very idea of crossing a hedgehog with a snake already deserves a separate and honourable Dr Schnobel award. "

Don't you think it's people who have already posed this challenge?

Since you're getting into mathematical finance, maybe you should start with the basics instead of making up incomprehensible stuff....

What mathematical programming is used for, I understand you mean the Belman equation.

In finance it is used, to find at a point in time all the optimal (according to the main problem) parameters of the system, using today's info. (Markov chains).

You can read the theory here https://en.wikipedia.org/wiki/Dynamic_programming (left-hand link in Russian).

As for your wave analysis, to be honest, I don't use such approaches, so I don't know how it works in practice at all.

I think that nothing good will come out of it.

The eternal question in all problems is not the tool with which the problem is solved, but what the model will be available at the input and how much it really explains the phenomenon of potential profit.

If you are interested in asset management using mathematical methods, I recommend taking a look at Markowitz and his descendants, and of course Karatzas Mathematical Finance.

 

Thanks, Quant, for the link on dynamic programming. Had some fun with the Fib calculation. According to the dumb algorithm described in the article, Fibonacci number calculation with number 45 (in MT4) takes me 381 seconds. Frankly speaking, I'm amazed (my processor is not the weakest but both cores are loaded 100%; everything is clear here - the dumb algorithm is recursive). I didn't dare to calculate Fibo with number 50.

Clever algorithm with memorialization calculates the same instantaneously.

Conclusion: no matter how many cores you have in your stone, you still can't do without protein brains.

Here are functions for both algorithms (function types and internal variables are declared as double to avoid integer type overflow). Those who have fast Intel stones may see for themselves:

double fiboDull( int n )
{
   if( n == 0 ) return( 0 );
   if( n == 1 ) return( 1 );
   return( fiboDull( n - 1 ) + fiboDull( n - 2 ) );
}


double fiboSmart( int n )
{
   double previousFib = 0; 
   double currentFib = 1;
   if( n == 0 )  return( 0 );
   if( n == 1 )  return( 1 );
   double newFib;
   for( int i = 0; i < n - 1; i ++ )
   {
      newFib = previousFib + currentFib;
      previousFib = currentFib;
      currentFib  = newFib;
   }   
   return( currentFib );
}
 

be simple, you can talk forever about those who are professionals in higher mathematics, but words cannot be taken out of the song:

as (already) many years of CHAMPI results have shown, not a single professional mathematician has ever made it into the first (second, third ... etc., the list goes on) of the top ten CHAMPI )))))

 
Mathemat >> :

According to the dumb algorithm given in the article. everything is clear here: the dumb algorithm is recursive.

"In mathematics and computer science, dynamic programming is a method of solving problems that exhibit the properties of overlapping subproblems and optimal substructure (described below). The method takes much less time than naive methods."

Read it carefully, and the second sentence too....

Description of the problem you found:

"To say that a problem has overlapping subproblems is to say that the same subproblems are used to solve many different larger problems. For example, in the Fibonacci sequence, F 3 = F 1 + F 2 and F 4 = F 2 + F 3 - computing each number involves computing F 2. Because both F 3 and F 4 are needed to compute F 5, a naive approach to computing F 5 may end up computing F 2 twice or more. This applies whenever overlapping subproblems are present: a naive approach may waste time recomputing optimal solutions to subproblems it has already solved.

In order to avoid this, we instead save the solutions to problems we have already solved. Then, if we need to solve the same problem later, we can retrieve and reuse our already-computed solution. This approach is called memoization (not memorization, although this term also fits). If we are sure we won't need a particular solution anymore, we can throw it away to save space. In some cases, we can even compute the solutions to subproblems we know that we'll need in advance.

In summary, dynamic programming makes use of:

Mathemat >> :

A clever algorithm with memoisation calculates the same thing instantaneously.

Conclusion: no matter how many cores you have in your rock, you can't do without protein brains.


Is the term wrong, by any chance? https://en.wikipedia.org/wiki/Memoization


Conclusion:

1. You don't need to make HUGE conclusions.

2. You should read more INTENTIONALLY or just read. Once you understand the concept, a new layer of methods will become available to you.

The code you wrote here: http://20bits.com/articles/introduction-to-dynamic-programming/ and many more "new" details.

 

1. Quant, I got the term wrong.

2. The conclusion is not hasty - and I demonstrated it with the code I gave above in MT4, not in any other language. There was no such code in your first link, but a pseudocode, as far as I understood.

as longstanding results of CHEMPIE have shown, not a single professional mathematician ever made it to the first (second, third ... etc., the list goes on) top ten of CHEMPIE )))))

budimir, three years does not pass as "perennial" by any criteria, so it's a bit early to draw such a conclusion. Regarding mathematics: I almost have to admit that for creating the system itself, the contrived mathematics is not very useful. A robotic system can indeed be very simple and not use any of the higher maths stuff. There is, however, an even more important area where mathematics is irreplaceable: risk assessment of a system and investigation (and proof) of its robustness. You can demonstrate beautiful balance graphs as much as you want, but without more or less rigorous mathematical justification of robustness these demonstrations are of no value. And the robustness checks of the style "make three dozen trades on the real account of your system, and based on the results I will decide whether to buy it" do not work.

 
Quant писал(а) >>

unfortunately, in my experience pilot + forex is hardly compatible....

Are you a pilot too?

1...5678910111213141516171819...21
New comment