Trading algorithmically a folio without stops (with IB), real $$$

Discussion in 'Journals' started by fullautotrading, Oct 16, 2013.

  1. AItrader

    AItrader

    Hi Fullautotrading,
    cc: Mo06, jcl366, smallstops, etc.

    Fullautotrading,

    thank you for your continuous updating on Gbot performances.
    It shows balls to honestly display the journal of the trading journey that you are making both during profitable and not profitable phases.
    You have said so since the beginning of the thread and you have stuck to it.
    You are keeping your word.
    Please keep us updated.

    Sorry for you guys ( Mo06, jcl366, smallstops, etc. ) but your latest comments are really poor ones. Is envy a motivation?

    If you want to criticise something, it is better to read all that has been written before about it.
    Second, post some detailed critics if you like but not just comments based on no reasoning, which you believe might be humoristics, but really show how poor is your comprehension of what you read or your lack of preparation in general.

    You may agree or not with the concepts of "avoiding information loss" that Fullautotrading has clearly detailed many times, and about the folios strategy in Gbot,
    BUT (yes it is a big but)
    if you want to criticize what Fullautotrading is doing you have to do it INSIDE or STARTING FROM (these are the inference keywords) his conceptual framework.

    It is poor logical reasoning to say that because in a trading approach with "information loss" (like stop-loss or profit-taking systems) something is done in a particular way (like WFA) then whoever is not using WFA is not doing something right.

    In fact constructive and useful critics for everyone to read would be to show (for example) that in a Without Information Loss trading system (like G-bot), the practice of WFA is useful. Is it true? Is it not?

    Dear Mo06, jcl366, smallstops, etc. put some hard thinking into your comments, that would be very useful instead of just posting Bart-like comments.

    To be completely clear: I am not saying that Fullautotrading strategy is better or not than another one. For sure it is a novel and different approach from a classic Stop and Loss / Signal based trading system.
    It looks completely different.
    Fullautotrading is showing us a system where there is:
    1) the concept of maintaining Information over time (not closing players until positive)
    2) of hedging, of scalping and so on
    3) all combined in several folios to be coordinated.

    Do we agree that it is a different approach to classic trading? I think so.

    Is it better? is it worse? I do not know but I am sure that the posting from Fullautotrading will shed more light on this interesting journey.

    Is it difficult to understand? You can bet, it uses so many new and different concepts that it is difficult to understand, explain, model.

    The only critic I can move to Fullautotrading, so far, is that Gbot is so full of novel trading ideas that it is overwhelming to understand.

    Guys, I like to be overwhelmed with such an amount of novelties and try to understand how all is working together.

    Again Fullautotrading thanks for your continuous effort
    And all the rest of you guys, please post information dense comments/critics that can help us all understand if a trading approach Without Information Loss makes sense or not and why.
     
    #291     Jun 27, 2014
  2. jcl366

    jcl366

    I have to agree that any system based on new ideas is a good thing.

    However, unlike the other trading journals this here has a commercial intent. There is nothing wrong with advertising a commercial product, but then you have to accept questions and criticism, especially when you dismiss competing methods as delusional. And when you react on questions and criticism with ad hominem remarks, don't be surprised when the tone of the discussion changes for the worse.

    I have been asked by a user on our forum to look into G-Bot and give my opinion, therefore my questions. The fact that I do not see how information of past trades can be used for future profits does not mean that it is not so. But as long as no mathematical or theoretical explanation is offered, only results can talk. Or have already.
     
    #292     Jun 27, 2014
  3. @AItrader

    >thank you for your continuous updating on ... performances.
    It shows balls to honestly display the journal of the trading journey that you are making both during profitable and not profitable phases.
    You have said so since the beginning of the thread and you have stuck to it.


    Thank you. I do think that integrity is the first priority. Then I might also lose a trading fight, but at least I have done so transparently and doing all that could be rationally done, under the given circumstances.

    >If you want to criticise something, it is better to read all that has been written before about it.

    Well some people are motivated by just the need to promote their business and exploit popular credulity.You will never have any useful feedback in that case. If one thing I have learned in my life, and you can use it as quote if you wish:

    Ignorance is invincible and there is no way to fight it. However, it is also self-defeating.


    > Guys, I like to be overwhelmed with such an amount of novelties and try to understand how all is working together.

    And you shall :)


    Let's recap a bit what we have been doing here, also for the benefit of those you cannot read the thread, or who do not want to read ("invincible ignorance fallacy").

    We have been proposing a new vision and approach to trading, which aims to remove all the elements which, from a scientific point of view, are nonsense and focuses on the activity of a scalping and hedging trying to maximize the effectiveness of this action. The key elements of this vision are the "preservation and use of the past trading information", which is a crucial element, commonly disregarded in other approaches.

    We have so far done, to use a metaphor, 2 "attempts of flight", and we are currently on solar day 152 of attempt #2, which is currently in progress.

    - Attempt 1. We started exploring a preliminary form of scalping/hedging games with a capital of about 200K and mostly using only a portion of it in our gradual exploration of instruments, mostly ETFs. This first attempt was characterized by an initial phase of DD which lasted several weeks, until we rose the PNL to about 25K. Then we had some technical problem with hitting the initial margin limit and a severe execution mishap with a single instrument (UGAZ). This incident caused a rather ugly loss on that instrument (alone) and, on the basis of technical problems, we decided to "pull the plug" while still in profit.
    The reason why we "pulled the plug" in that case was the realization that changes where needed to deal with the order rejection issues. (At that time, I clearly stated that the interruption was a "logical" violation of the trading information preservation and use principle). However, it made sense in the perspective of the journey we are doing and the necessity to have a clean restart. So we made the changes and started "attempt of flight" #2.

    - Attempt 2. This is the trading session currently running and we are now on day 152 of it. This includes holidays and weekends. In this session we started with more mature scalping/hedging game and we continue to improve them during the entire session. Our investor who is generously supporting this research, considered the previous mishap and the problems with too small capital, decided to throw in another 170K. Thus rising the available capital to about 370K.
    This session has also started, like the first one, with several weeks in drawdown (which I prefer to call "investment phase"). Note that initially I have been using only a small fraction of the available capital, mostly using at most 1/2 for margins. After the "investment phase", in which anyway the minimum drawdown was always contained within 10/15% of margin usage, we reached a peak pf PNL of about 31K (we have currently executed about 12,000 orders).
    After this peak the folio has started another larger "load" phase which is currently in progress.The alternance of these phases are quite normal in real world trading, as of course you alternate an investment phase to a phase where (hopefully) to cash in the results of such investment. This new load phase has been quite useful because has indicated clearly the necessity to appropriately size the "protective" orders based on opposite side players (note that this is a form of usage of past trading information).
    The circumstance that we were "caught" by this new investment phase while testing multiple 3X instruments on metals has also urged to improve the mechanisms of hedging across layer, that is the use of instrument correlation to maximize the hedging action. It has also clearly indicated some challenges dealing with leveraged 3X instruments.

    The main aspect of this challenge is that while in an ideal world of instruments which behave say like some symmetric stochastic process, we could very easily apply all the beautiful ideas about hedging (both on the same layer and across layers), with these specific instrument (leveraged 3X) which have strong decay, the challenge is to both to hedge, but at the same time keep the short imbalance (that is why we are using the "Bias" game with short only constraint).
    So there is a "conflicting task" of, hedging (with superposed buy orders), but at the same time keep the long term short unbalance. Clearly it's necessary for these instruments to remain unbalanced, because their intrinsic drift is larger than what any scalping engine could ever make. So it's essential not to have the drift working against you in the long term. I also recognized that in the folio we had too many metals, and with extremely volatile instruments (3X). We sort of "got caught" in this situation while testing the various instrument, and right in the middle of replacing some instrument which were become impossible to trade algorithmically (like JDST, DGLD, ... and various other constantly not shortable).
    Anyway, given this situation, the challenge is always to be able to deal and recover through hedging from any situation one can have while trading.

    Assuming that difficult situations will never happen to you, can be left to realm of Magical Thinking, where you have those magic "signals", which make your equity curve go always up. Unfortunately, I am a scientist and don't know about that, but if interested you may contact the experts and gurus who have bothered to manifested themselves in this thread, and of which the web is also full.

    It's actually when you have an instrument "running away" that there are the greatest moments of learning from the perspective of tuning the hedging actions and game improvement if you take control of the situation and resolve it, instead of quitting like a clueless cry baby. It's also true that, paradoxically, a running away instrument will often also provide the opportunities for the largest gains you can currently make in the folio. As of course you can use some of the running away move for you advantage, and that gain is usually larger than any other gain you can currently make in the folio.

    So unless, you did not know what you were doing from the beginning, quitting on an investment phase is actually the most nonsensical action you can do, from a rational point of view. Clearly, different would be the situation, if the investment itself were nonsensical, like, for instance, if the position on NUGT and USLV we currently hold were long. In such a case, you would be also fighting against and adverse structural long term drift. And, of course quitting, would be just the late resolution of something which should never have been started in the first place. So it's important to see the large picture and where you are within it.

    If you dig a bit, you soon realize that our reality is all based on information. And the use of the past trading information, along with the information about the structural characteristics of traded instruments is all that matters.

    The conception of mkt giving out predictive "signals" for you to have a statistical edge, it's something which I as scientist must reject as a naive form of "magical thinking". In addition to not existing any statistical evidence of that, it's obvious that it cannot be so, as that would be cause of market self-defeat, and this is supported often by people who are fully aware that is nonsense but, nevertheless aim to exploit the credulous or the desperate (just like lotteries do).

    We cannot simply stare at a running away instrument doing nothing. But we must hedge a way out. Nor we can "stop and forget" it - if our investment makes sense (if we need to stop it, the investment did not make sense in the first place) - but we need to have the mechanisms to smoothly hedge out of the situation.
    And that is all there is to it. You must be able to hedge out of any meaningful investment. If you simply "quit" just in the moment you are actually given the possibility to realize the largest gains (stemming from hedging action on that instrument running away) you are simply not being rational.

    Also thinking that you will never had to hedge or confront yourself with PNL at -10%-15% (or much more) of the capital and be able to deal scientifically with the situation, is strongly delusional, and unrealistic and only proves absolutely no experience with real trading.

    It rather belongs to that sphere of magical thinking which when we are kids all like to believe, but that we must also learn to watch as a childhood memory when we mature in our cognitive processes.
     
    #293     Jun 28, 2014
  4. However, It seems more pronounced in the returns of the assets

    as in the returns follow a fat tail distribution function with a drift appoaching 0 the shorter
    the time frames. Has anyone else noticed this ?
     
    #294     Jun 28, 2014
  5. Mo06

    Mo06

    Fullautotrading, you do some across as somebody who prefers to use 150 words, where perhaps four would do.

    Trading is very simple.

    From your posts, you appear to be a losing trader, who thinks that if they just keep repeating what they are doing, they will somehow become a winning trader.

    Anyway, good luck with that.
     
    #295     Jun 28, 2014
  6. "if all you have is a hammer, everything looks like a nail"

    And, we are still waiting for your journal.

    Let us see those "signals" at work, enlighten the readers on how the mkt communicates to you in person its "intentions". And, please, with more than 1 order a day, and hopefully with more than 2K capital, to have some hope of some statistical significance before we die.
     
    #296     Jun 28, 2014
  7. Mo06

    Mo06

    "And, we are still waiting for your journal."


    124 posts to date. :cool:
     
    #297     Jun 29, 2014
  8. Referring to this: http://www.elitetrader.com/vb/showthread.php?t=255007&page=18 ?

    LOL. That should be fun.

    Please, let me know when your "Summary Weeks" will show that you are in a "through" of your equity curve (well, I know your equity curves have only "crests", but nevertheless, just in case you decide to give it a touch of realisticity and not give the impression you handpick what is convenient to show and make "selective reporting"), so I can come like a vulture and kindly encourage you to liquidate your position :)

    Also, in that occasion, I'd like to know precisely how your hedging techniques work (well, I know you need none because you always "guess" right).

    I would also like to inform you that the James Randi Foundation has a prize of $1 Million for psychics (http://en.wikipedia.org/wiki/One_Million_Dollar_Paranormal_Challenge) and that you should definitely apply. You would get it for sure.
    1M would surely be useful to give some statistical significance to your trading activity. (What is the capital being traded please ?)
     
    #298     Jun 29, 2014
  9. Mo06

    Mo06

    "Please, let me know when your "Summary Weeks" will show that you you are in a "through" of your equity curve (well, I know your equity curves have only "crests", but nevertheless, just in case you decide to give it a touch of realisticity and not give the impression you handpick what is convenient to show and have "selective reporting"), so I can come like a vulture and kindly encourage you to liquidate your position :)

    Also, in that occasion, I'd like to know precisely how your hedging techniques work (well, I know you need none because you always "guess" right)."

    What exactly is an equity "through" ....?

    And "realisticity " ...?

    As I have stated, my journal is updated at the end of each week.

    I do have drawdowns, but my equity curve is one which rises from the left to the right hand side of the chart, otherwise what would be the point ?

    I don't always get it right, and if you read the journal you would already be aware of this.

    I very rarely hedge.

    I don't trade indices.

    I don't trade FX.

    I don't trade options.

    I don't claim to be a great trader, far from it, and I have made many mistakes along my trading journey. But I do believe that I have learned a thing or two about this game.

    Good luck in your trading.

    I will be posting the next update at the end of the week.
     
    #299     Jun 29, 2014
  10. What exactly is an equity "through" ....?

    http://en.wikipedia.org/wiki/Crest_(physics)

    >And "realisticity " ...?

    http://glosbe.com/en/en/realisticity

    >I do have drawdowns, but my equity curve is one which rises from the left to the right hand side of the chart, otherwise what would be the point ?

    I agree. There would be no point.

    >I don't always get it right, and if you read the journal you would already be aware of this.

    You are still eligible for the $1M cash price. For James Randi it's enough that you prove to be statistically significant. With your possibilities, why give away the $1M ? It's not rational. (Well of course in that case they would not allow you to do any selective reporting, but that should not be problem, right?).

    >I very rarely hedge.

    Why should you hedge ? I would also not hedge if I were you.

    >I will be posting the next update at the end of the week

    Still waiting to know what is the capital being traded, the margin used and the margining method.
     
    #300     Jun 29, 2014