Log in

Artificial Bee Colony algorithm

By KAUSTUBH 4307 days ago Comments (10)

ABC algorithm is one of the latest NN algo and on bigmike site, effors are on to develop autotrading strategies based on this. I thought traders here would be interested in this.

Sharing here some links about it.








I have noticed similar NN algorithm with wave59 called 'Hive technology'.




  • fxez 4306 days ago

    I couldn't see the link at big mike without elite membership. But I'm sure you already knew that. I would be more interested if you shared a public link.

  • KAUSTUBH 4306 days ago


    The bigmikethread is in their elite section and requires elite membership to view it. But I am sharing other links and the idea here which they have posted.

    The concept is beyond the grasp of my brain, but I thought we have traders here like JohnLast who would be interested in this kind of stuff.

    I am reading their thread and will update the method about how they use the algortihm in trading.


  • jaguar1637 4305 days ago

    As far as I can understand, it looks like the Particule_Swarm Optimization, based on birds instead of bees.

    I wrote something about this, I was stucked in the calculation of the rosenbrock..

    Well, If I understand the mathematical process, what provides PSO, is the vector of the waves. The indicator I wrote, shows the bird who is the true leader of the flight formation

    So, Birds Or Bees to follow ? this is the question

  • jaguar1637 4305 days ago

    BIRD SIDE ;)


    The solution is Particle Swarm Optimization which is a type genetic GP or evolution study thus the EA indicator is called PSO
    The version above is rushed only 10 min training. The final version will take several hours.



  • JohnLast 4305 days ago

    Thanks for the links. I was thinking that those algorythms would work as the MT4 optimizer. For example for open office there are two interesting algorythms for non - linear optimization (of course everything is free). 

    DEPS Evolutionary algorythm

    (dierential evolution particle swarm) is a hybrid multiagent approach based on the
    particle swarm optimization paradigm (Zhang and Xie 2003)

    SCO Evolutionary algorythm 

    (social cognitive optimization) is based on social intelligence and human learning mechanisms (Xie et al. 2002).

    I think (consider) the non free lunch theorem to hold true.

    "To make matters more concrete, consider an optimization practitioner confronted with a problem. Given some knowledge of how the problem arose, the practitioner may be able to exploit the knowledge in selection of an algorithm that will perform well in solving the problem. If the practitioner does not understand how to exploit the knowledge, or simply has no knowledge, then he or she faces the question of whether some algorithm generally outperforms others on real-world problems. The authors of the "(almost) no free lunch" theorem say that the answer is essentially no, but admit some reservations as to whether the theorem addresses practice."

    What really matters is how to exploit the knowledge of the problem in order to select the algorythm that is able to solve the problem.

    That is a kind of metaknowledge and it is a dark science regarding the markets.


  • JohnLast 4305 days ago

    I think it is hard to search for the ultimate algorythm. Before that it is advisable for practical reasons to focus on one algorythm and to understand it really well how it behaves regarding the markets.

    Ultimate algorythms do exist and they are related with quantum computing. For more information about this you can ask Alexandra (but high end profiles in applied mathematics is required - no trading applications at all ;). ).

  • KAUSTUBH 4305 days ago

    Thanks for the comments....We have had SSA and wavelet algorithms and now this....SSA has been well implemented for MT4 and many nice indicators based on SSA coded by John and Jaguar. On Bigmike site, they discussed about ants, birds and bees and finally settled on bees behaviour. There is link to algorithm in C and java here if anybody is interested. I will be updating the thread and am trying to understand the concept and its application to markets. At the end of the day, the PL is what matters.

    some more links http://www.bigmiketrading.com/ninjatrader-programming/21062-experimental-one-step-aforge-based-neural-networks-predictor-open-high-low-close-data-neural-network.html


  • KAUSTUBH 4304 days ago

    A link to free AI course from Stanford.



  • jaguar1637 4044 days ago

    Well, this is very interesting, bcoz I just wrote the particule swarm optimization.

    This is in the same spirit ants, bird, bees , ... what else ? pingouins ? Platypus ??? lol

  • JohnLast 4043 days ago

    I was readin some days ago about the new quantum computer: The D - wave systems. It ises the principle of quantum annealing. It is interesting how the computation principles are derived from the nature itself.

    And that is derived at different levels.

    I think I will write a separate blogpost on this topic.