Mathematical algorithms?

Discussion in 'Trading' started by Kastro_316, Jan 14, 2006.

  1. How do they apply to trading? Do they help with steps/processes related to programming a trading model?

    I did not learn these, so i would like to learn a little about them. Any information would be great :)

    Thanks guys
  2. As the space of mathematical algorithms is infinite, you put yourself to a formidable task.

    Basically, like many other decision processes, the final outcome is binary: to pull the trigger TRUE or FALSE. Typically such decision processes are broken down into substages involving sometimes many intermediate variables being derived from the primary observations or inputs. In fact our popular indicators are good examples of these.

    How to do this is mostly not obvious. The multistage decision process is often an attempt to tackle an imperfectly understood process by approximate means.

    These intermediate variables must be combined somehow into your final, often binary outcome. If you want to fully automate your decision process like in a military fire-control system or a market ATS, you will have to come up with a formula to do this. Discretionary traders would rely on their instant decision power to do this. Few discretionary traders would work with raw observations exclusively. They will likely also fall back on some computed intermediate decision variables, i.e. resulting from algorithms applied to primary observed variables.

    In general, learning about these is demanding. In markets you will be pitted against the best brains money can buy. This is confirmed by the popular but uncontested belief that 95% of those that try their hand at it go without success. It should also be obvious that working strategies or "algorithms" are not easy to come by.

  3. A similiar question was asked a while ago ... go through the replies for some general industry studies.

  4. How would one learn of this power? :)

  5. Feel the Force, Padwan.

    May the force be with you.