Everything you ever wanted to know about developing systems but was afraid to ask

Discussion in 'Strategy Building' started by Murray Ruggiero, May 17, 2012.

How do you get most of your system ideas

  1. Magazine articles

    1 vote(s)
    2.4%
  2. Things I see on charts

    9 vote(s)
    22.0%
  3. Combine classic rules and indicators I see in books

    2 vote(s)
    4.9%
  4. Develop and test a theory then build your system based on those results

    29 vote(s)
    70.7%
  1. Many here are looking forward to this tutorial on system development. I am sorry for any family illness that may have befallen you.

    On the other hand I notice that recently you have been making almost daily posts to at least one other thread. This thread was started 2 months ago, contains 10 posts - 6 of 10 by you and the only system development wisdom I can find here is that traditional indicators don't work as system inputs.

    All due respect Murray but you started this thread 2 months ago, committed to post content at least several times a week and illness aside have been posting regularly on other threads - so where is the beef?

    That price based momentum and stochastics are not very useful in system development is hardly big news to anybody. If you would please post something that is not already widely known or even slightly useful I would be most grateful and will be happy to publicly thank you and sing your praises.

    jack
     
    #11     Jul 18, 2012
  2. Murray Ruggiero

    Murray Ruggiero Sponsor

    When developing trading system there are many issues to address
    1) Data
    2) Selecting what to trade
    3) A good premise
    3) Understanding the science behind what you are doing
    4) converting a premise to code
    5) Testing your premise
    6) Judging system reliability.

    Here a subset of issues we will start discussing next week. We will start with discussing data.

    Here a preview, on a market by market basis how you roll futures contracts could make a 30% difference in results!.
     
    #12     Jul 20, 2012
  3. The poll topics were interesting and spotty.

    What I do was not included.

    The most "popular" answer doesn't work when a corrolation is made with these responder's performance results.

    Any system of trading is measured by how closely it performs in extracting the market's full offer.

    So, I conclude that stating the opportunity (Problem statement) comes down to three principles:

    1. Extracting the Market's full offer,

    2. Knowing completely the operating pattern of the market, and

    3. Processing information using the mathematics of the Marketplace.

    By making use of the three principles, a person can deduce (using Science) how to manage the information flow, deploy the proper tools as an extension of man, and Personally operate with a routine that allows the full offer of the market to be extracted.

    No one is ever, for any reason denied the opportunity to extract the full offer of the market.

    Anyone can concieve what the transition curve is and how the curve becomes asymptotic to the full offer of the market, continually.
     
    #13     Jul 22, 2012
  4. SamGold

    SamGold

    Do you have an irresistible urge to spray your babble-diarrhea continually?. Interesting pathology.
     
    #14     Jul 22, 2012
  5. To examine magine articles, It is probably best to examine publishers and the purpose of magazines. I am given magazines so the publisher can say I read the publication. He may be wrong, however.

    I decided to outline and to write five books. It took me 48 hours to have a senior editor at the first company I called. In college I was offered a Writer's Agreement with a major publisher. Book publishers have no standards and will publish just about anything.

    I looked at the classic development of the financial industry from the conception of trading to 1957. About 100 new ideas occurred during those centuries. In modern times, these ideas have been tested. From the earliest of times the pertinent ideas had come to the fore for taking the market's full offer. So I use this approach to trading and I achieve the market's full offer. It seems there is nothing new. (See a book by Dodd and Granville, for example).

    "Develop and test a theory then build your system based on those results".

    Since a system was designed and developed long long ago and it works, why would I go through this type of described journey.

    The market is not theoretically based. It happens to be such that its operation is deducable using science and mathematics.

    The foundation of the market's operation is the mathematics of its operation.

    Anyone can look around and see what is NOT being used mathematically. The reasons that eliminate each sphere of mathematics is quite clear.

    All scientists gong into a given field, early on, inform themselves of the applied mathematical field involved in their planned field of endeavor.

    The second thing they do is get very clear and straight the dimensions of the variables that are involved.

    This was done quickly following the DOW theories to get directly and correctly right back on the topic for making money.

    The foundational people who developed the mathematical field of endeavor for the markets were Keynes and Carnap.

    When I studied theoretical physics under the auspices of IBM who did not settle for anything less that the best professors, nobody, but nobody, saw any connection to financial markets. Times have not changed.

    Now you have some informed input from a trading practicioner who is using the correct mathematics and whose has a complete descriptive system of the market's operation. I also have a complete syste that gives me the totality of all the signals for trading to make money to extract the full market's offer. Note that two systems are involved.

    In modern times, languages have been invented for using any size computer. The language that fills Keynes' and Carnap's requirements is the RDBMS. A familiar language is SQL. Basically data is flowed into tables and from these tables new degrees of freedom are created and put into other specialized tables.

    The first deduced pattern of the marketplace, by deduction again, has to repeat. It is the nature of a pattern to have three parts: a beginning a middle and an ending. RDBMS are of a nature that they "reset" at the appropriate event (the beginning of the overlap of the pattern).

    The granularity of the markets dictates (to the reasoner who is doing deduction in a scientific manner) that the pattern of the market MUST overlap.

    In market analysis, you cannot choose the improper variables. Most of the published work makes this mistake and therefore cannot be used to further understanding of markets. Dodd and Granville did not make this mistake.

    There are two basic variables and they are NOT opposites but are orthogonal. They happen, by their construction to take Time as a variable OUT of the mathematics of the markets.

    All the above is usually rejected by people who are never able to extract the full offer of the market as it is continually offered.

    Strategy system design is a forum that is fairly unrestricted. Be careful out there.
     
    #15     Jul 22, 2012
  6. Thank you for your questions.

    You post where I post, so you may want to reflect on your question and see if you can answer it yourself.

    Take your time It is only as urgent as you make it) and let us know how you feel about your posting in paralle (but lagging)l to me.

    Do you have a doggy or a kitty? Do you give it (them) prozac regularly? Check with your vet if you have a problem with them seeking identities in your mind.
     
    #16     Jul 22, 2012
  7. You are welcome, Jacky. You are welcome... As always.
     
    #17     Jul 23, 2012
  8. Murray Ruggiero

    Murray Ruggiero Sponsor

    Let's start talking about data.
    Understanding how data effects testing and developing trading systems is very important.

    One example is how futures contracts are rolled over. There are many ways to do that. First rolling on volume , open interest and then volume and open interest. We select the most active contract. Another way is to roll on fixed dates used by many commercial traders, for example 30 year treasury futures roll on the 26th day of the month before expiration. How you roll makes a big difference on a market by market basis as much as 35%. If we trade a large portfolio with the same system these results balance out to maybe a 10% difference.

    Another issue is how the birth of the electronic markets effects long term backtests. Some data vendors have electronic only data and combined electronic /Pit history. This combined history is not good because it include old history prior to 2008 , when many electronic markets were not liquid.

    In order to create long backtests we have created, Pit/Pure electronic data series rolled on fixed dates. We create a roll from pit to electronic when the volume of the electronic markets became greater than the pits. Pinnacle data create these back adjusted contracts with my help and make them available.
     
    #18     Jul 24, 2012
  9. I noticed that many stocks (and the SPY) got much more volatile around late 2008, on an hourly basis. I never looked into it, but I thought it could have been some fundamental change in the markets (electronic markets?) or some big players went bust, who would have otherwise played the range.
     
    #19     Jul 25, 2012
  10. Murray Ruggiero

    Murray Ruggiero Sponsor

    I think high speed traders do have a effect on why we get more intra-day volatility. I think you will see if from 1 hour down to tick level, You might even see this effect for on as long as 90 minutes. I think it has some but less of a effect on daily data. The change is the markets will make multiple moves in the same day we we don't see on daily bars.
     
    #20     Aug 1, 2012