null - Page 3
Page 3 of 8410 FirstFirst 12345 ... LastLast
Results 21 to 30 of 93

Thread: null

  1. #21
    Even When You Need to Win, Most Will Eliminate

    If you haven't Previously, https://papers.ssrn.com/sol3/papers....act_id=2856963.

    Below is the Abstract
    What would you do if you're encouraged to play a sport in which you were given $25 and permitted to place bets for 30 minutes on a coin that you were advised was biased to come up heads 60% of this time? This is exactly what we did, gathering 61 quantitatively women and trained men to play with this game. The results, in summary, were the majority of these 61 gamers didn#8217;t place their bets demoning a panoply of behaviorial and cognitive biases. About 30% of the subjects really went bankrupt, losing their25 bet. We discuss optimal egies, evaluation of the opportunity to play the sport and its similarities to investing in the stock market. The impliion of the study is that people will need to become trained and edued in how to process decision. If these quantitatively trained gamers, playing with the easiest game we could think of involving uncertainty and favourable chances, didn#8217;t perform well, what hope is there for the rest of us when it comes to enjoying with the biggest and most crucial game of all investing our savings?

    The daring announcement says it all really.

    In the this study, you have $25 to perform an a bonded advantage of 10% if you simply bet heads. With the timeframe, about 300 bets could be made by one. If you just bet on heads each time, it would be almost impossible, if not impossible to reach the maximum of $250.00.


    Each and every individual KNEW this going in, they were informed, no matter what, betting heads gives you a 10% edge. If you were able to find/create this scenario on the market, that is THE holy grail...

    A brief summary of the outcomes:
    Quote Originally Posted by ;
    Remarkably, 28 percent of the participants went bankrupt, and the average payout was only $91. The maximum was reached by only 21% of those participants. While two-thirds gambled on tails at some point in the experiment, 18 of the 61 participants wager everything on one toss.
    These results are amazing!
    1. Nearly 1/3 of those players went bust with a edge. Imagine whether casinos had edge that is 10%. These people had almost a 10x higher edge than a casino, and almost 1/3 withdrew their money and dropped...
    2. Only 18 of 61 people attained the maximum of $250.00. Basically everyone should have attained the maximum if they just followed the system.
    3. And, this is the most fascinating, almost 2/3 of those players really bet on tails at times during the experiment... Almost 66% of those people took stakes they KNEW had a 10% edge against them.


    This only shows, even if you're able to win and have the system to triumph, most people will still lose. Most people drop with a guaranteed 10% edge... You can bet your ass most will lose with a smaller edge that is dynamic as the market changes.

    Why? People don't do what they are supposed to, even if they are aware of what they are doing will not do the job, or is wrong.

    Even when given a sacred grail, 1/3 of those people will drop, and only 1/5 will find the maximum payout. Even if you mathematically shouldn't lose (possess the holy grail), a majority of people will still lose... Along with this, 1/3 of most people will make bets that don't have any positive expected values, even if they know this to be true.

  2. #22
    Rolling Over to the Immediate Practical Future

    -- This first section is much more of a small upgrade. The first image will be referred back to throughout the main portion of this post. --

    You know that it's going to be a good weekend when it begins like this:


    Going to spend a lot of time this weekend studying some data mining prejudice stuff and working with a few of my systems.

    Specifically will be working together with Walk Forward Analysis in hunt of 2 things.
    1. To see the frequency distribution information gathered in sample, orders the out of sample behaviour of the market.
    2. To observe the frequency distribution information gathered in sample is overstated or understated.



    I always like recommending books, and for those seeking to learn a lot about actually taking a deep dive into technical analysis, I'd check out the book under.



    I heard of this publiion from Relativity and I can honestly say it's worth getting. I find myself flipping back through it when I am testing new ideas.

    -- Main Idea of Post Below --

    I have a new book coming this weekend that's about finite and infinite games... Sounds awesome I know, and even though it has nothing to do with markets, randomness, etc, I've a feeling it will give me a lot of new thoughts that are applicable to the market.

    One thought in particular I've been really milling within my head is considering rolling over trades to new X Axis Stops in order to extend a finite round a little longer. The reasons I am seeking to do this is so I can get the time that is optimal push existing places as far as they will go and to incorporate into places.

    There are really two ways of moving about rolling a trade into a newer A Axis prevent that's further away:
    1. Roll into a higher time frame move by using timeframes that are multi - This is what is usually discussed about. I will look into this second.
    2. Roll into a new window at the time frame. - This is what I am considering.

    The reason I am very curious about #2 is that. When rolling into higher time frames to extend duration, you're extending your volatility risk (when I say this, I mean the range of anticipated moves. Case in point: 1hour ADR vs 1D ADR, should you cease was 1ADR and you also rolled into a 1D ADR trade, you would have to expand your stop to fit this higher time frame together with your target if you're looking to acquire the expected results of this Daily Data. I hope this makes sense.) . Yes you're rolling into a higher time frame chart so that your target will truly expand up, nevertheless your stop will also widen quite a bit (frequently more than twice, at least with my system), even when you're still riding the most important direction of this market.

    Should you roll into a new window at precisely the exact same time frame, you're still rolling with main direction, your duration gets extended, your volatility risk stays the same (take the preceding example, your ADR doesn't change even though the trade has rolled over to speak), your goals expand since your lower bound ADR would be lower because of the drift, along with your stops would shrink since the upper bound ADR would be lower because of the drift.

    I am thinking these rolling periods will be great methods to milk present trades, and add into new trades.

    That is kind of what I did using the EURJPY trade I posted.

    By way of example, when I first entered was risking 30 Pips to create 80. To make things easy, lets assume for every place 1 pip =1.00

    Round the next entrance, I rolled over the first trade into a new window. Adjusting my existing stop and target changed my risk so I was risking, 11 pips to create 80. Then I add after the roll over, and with this added leg I am risking 20 pips to create about 70.

    So because I rolled the trade to a new window at precisely the exact same time frame, instead of a higher time frame. I was able still trade using the very same probabilities I was using when I put on the first place, except today I am risking 30 pips or $30.00 ($11.00 on the very first place, $20.00 on the next) to create approximately 150 pips or $150.00($80 on the first place, $70 on the next).

    By rolling we can double reward while maintaining our risk the same.

    To best describe this kind of window I am talking about, I will refer to some other image in the book I published above.



    It is essentially continuing to roll into the immediate practical future in order to maximize profit potential and reduce risk. Trading like this can feel counter-diminutive because humans are programmed to optimize the chance of gain, over their gain's potential.

    Example: Suppose that the initial probabilities we are trading using are 60%, so when we enter the very first trade, we've got a 60% of winning 80 pips and a 40% probability of dropping 30.

    Before we roll into the next trade, the probabilities of their very first trade being a winner are higher than 60% since price is nearer to our first target.

    Once we roll our trade into the immediate practical long run, the probabilities to our net position go back to 60/40 only today we are risking 30 pips for 150. We actually deoptimized that the chance of gain (which is exactly what people, by default, do not want to do) in order to optimize our potential gain (which is also exactly what people, by default, do not want to perform).

    NOTE: it's crucial to confirm your windows are independent from one another, and if they are not, you may use the calculation from the A Correlation Problem post to figure out what your new chances would be after rolling.

  3. #23
    Shing Probabilities

    Consistently love the Beginning of a new week.

    Working with a few walk-forward analysis material this weekend turned out to be quite effective, and I found a few interesting things.

    The very first thing I noticed when doing the walk-forward analysis is how the probabilities change over time. In this informative article, when I say probabilities, I'm specifically speaking about the Non-Linear Trend probabilities which can be seen, and I've written about on this thread. When you look at one huge chunk of information and gather your data, you don't really see the whole image of those numbers came to be, and also what it may look like when these numbers change. Together with the ahead analysis, breaking the one sample that is huge allows me to see how the Non-Linear Trend probabilities changed over time.

    So I started looking at three different window sizes to gather the probabilities of price peaking to the upside or downside during that window. The window sizes are selected. Keep to the instant future that is practical.

    I picked 12, 48, and 500. I looked at 10,000 candles, used a sample size of 100 samples of every window to figure out the probabilities. I have attracted a normal bell curve over every chart to understand how the probabilities should be dispersed. Doing this study has direct me to separate the Trend part into is own separate indior I call NLT. The goal of the NLT is to determine leadership.

    NLT12


    NLT48


    NLT500



    The NLT12 probabilities follow a Fairly normal supply. So that we could see that the probabilities do skew to the more than normal this is looking at EURUSD.

    The NLT48 probabilities also comply with a fairly normal distribution, except this time that the skew is considerably more conspicuous.

    The NLT500 distributions get interesting. This supply starts to follow the same supply talked about in this informative article. The probabilities wish to stick to the edges (tops and bottoms), whereas using all the bigger windows, the probabilities wish to follow a more normal distribution.

    Now what I'm really looking at here is the trends of the probabilities. The reason is, we know the probabilities are likely to contract back towards 50/50 , even though the probabilities to an up peak may be 60% over 48's window dimensions. You'll be crushed even though the probabilities are technical in your favor at the moment, if you are trading in this arrangement. However, like all things in the market, they change over time. On the flip side of that, if you're able to follow this contraction and transaction accordingly, before it starts, you can avoid the drawdowns that arrive with fad changes, and also get ahead of the trend.


    I think it's important to know how they change over time so you are able to exchange the current probabilities but also anticipate how the probabilities will be dispersed in the instant practical future.

    TL;DR: Can you rather wait for the probabilities to move from 60 percent for up peaks to 40 percent before searching for price to make a downward peak, or would you rather begin fishing for down that peak as the probabilities start dropping from 60% to 40 percent. I know what I do.
    I think it's important to know how they change over time so you are able to exchange the current probabilities but also anticipate how the probabilities will be dispersed in the future according to their frequency distributions.

  4. #24
    Four Pillars of a Technical System
    There is a trader around here and on other forums that goes by a few distinct names. On here he is iDouble. He is also on other forums under the name SignalBendergod and god knows how many other names.

    Some folks think he's full of crap, others highly revere him. And the others fall somewhere in between.

    IDoubleStochs includes a really interesting thread according to an indior he calls the B2i. The indior is simple, it is essentially a custom-made pivot point that's returned in 1 phase 72 percent of the time, much like a pivot point.

    I am not really too interested in his humorous looking rectangles. What I am interested in is his ideas on systems in general.

    He talks about all genuinely powerful technical systems are based on what I like to call the Four Pillars. They are:
    Time
    Management
    Magnitude
    Probability

    I like to carry it a step farther and apply probability to the 3 additional pillars Too. I want to understand the probabilities associated with nearly everything I measure and the way the probabilities are distributed.

    Like I said in my last post about the NLT. The intent of that indior is to focus on direction and the probabilities. It has a great deal to do with timing. Where things get a little tricky that's, whenever you are focusing on one of these columns, they often bleed into each other.

    The following indior I dip into on here will probably be focused on magnitude. And time.

  5. #25
    More on Changing Probabilities

    Quote Originally Posted by ;
    Changing Probabilities Always love the Beginning of a new week. Working with a few walk-forward analysis material this weekend turned out to be productive, and I discovered a few interesting items. The very first thing I noticed when doing the analysis is how the probabilities change over time. In this post, when I say probabilities, I'm specifically referring to this Non-Linear Trend probabilities I've written about on this particular thread, and which can also be observed on...
    Broke down another section of my dashboard into its separate indior. This indior is specifically centered on the probability and time columns. Essentially provides the probability of levels being hit.

    It is again interesting to see how the probabilities change when the window size varies. Smaller windows, the probabilities tend to accompany distributions with skews that are slight. Larger windows, the probabilities tend to follow an distribution.

    OLS12


    OLS48


    OLS500


    Again, if you can anticipate the way the probabilities will change, you can use these changes to your benefit when it comes to time the market.

  6. #26
    Revenue Requirements

    Something I've observed just two traders really touch on is that this idea of determining revenue requirements and the necessary environment BEFORE even touching the market. Those 2 people are iDoubleStoch and Jarratt Davis (I've written about a lot of his job at my Fundamental Trading thread).

    The main points from each individual are essentially the same however. Determine your earnings needs and what surroundings is needed to meet these requirements. If an inherent doesn't provide an environment to reach your earnings needs, then you've got a few options:

    1. Loe a new inherent.
    2. Find the difference to be made up by extra.

    Coming up with the earnings requirement is easy and usually is centered around your budget/how a lot of the budget that you want to be paid for with FX profits.

    When I first started, my earnings requirement was $8,000/month. This included all of my invoices, amusement, etc. and left about $2,000 left over each month to re-invest into an IRA, trading account, additional chances, etc..

    Side Note:
    A lot of folks prefer to state that they will re-invest everything into the fx account, and compound to the moon and back. This never works like this. I also find immense advantage in pulling a certain % of profits outside each week and spending them. This way you can truly realize your job is paying off in the brief term.

    I also would imply, and this is my personal opinion, re-investing your profits into other ventures than FX. I, for instance, LOVE real estate, foreign and domestic stock indexes, and commodities. Pretty much every excess dollar I've enter one of these 3 bins. Usually the first bin. Nevertheless, the major issue is to create flows of revenue.
    -- End of Side Note --

    So today you've got a sales requirement, next let us talk about what sort of environment is imperative to make this happen.

    The second issue to check at is what sort of consistent performance and account balance is needed to realistically meet this objective. The part of this segment is to be REALISTIC.

    Yes the math says that you can make $8,000 a month with a $100 if you just compound to the moon and back, but realistically this could be almost impossible, and if you could do it , it could be so plogically taxing you wouldn't be able to perform it more than a couple of times, and you definitely wouldn't be able to perform it on any consistent basis.

    When I first began, I said I had an account balance of $12,000-$16,000 and a consistent monthly profit of 40-60 percent to realistically meet these goals. For me, a month making 40-60 percent is very realistic. I attempt to shoot for returns beyond that, I experience drawdowns and run-ups that are plogically hard for my pallet.

    Thus, we now have our earnings requirement, and part of the necessary environment to meet that requirement (the part we restrain at least).

    Next, we need to consult Mr. Market to see if he's got an environment that meets our requirement. Really where we are currently handling things that are beyond our control, that is, and that is where things can get hard. You need to be great at collecting and assessing information to do this right.

    You need to figure out which inherent assets produce the hourly/daily/weekly/yearly moves required by your platform to meet your revenue needs, together with the first half of the surroundings that you set.

    Your machine may win 100% of their time, but if it only provides you one trade a month, then you won't be able to consistently reach your earnings requirement, particularly factoring in how matters like correlation, number of occurrences, and a few other subjects I've written about in this particular thread.

    For instance, my system at the time had spent durations of 24 hours or less, and each trade gave an expected value of 30 pips after it is all said and done. Let us say we're trading with .1 lots so 30 pips $30.00. In order to meet my monthly earnings requirements, I need a mean daily requirement of 400 pips of expected value to be set on each day (5 days a week with 4 weeks a month).

    Therefore trading 1 inherent was not a choice for me. It might have been impossible to meet my earnings requirements. In order to have fulfilled them, I would have to exchange 13 inherent daily.

    That may be a lot to manage, so if we want to change that, we've got a few options:
    1. Loe a new system - this isn't recommended by me.
    2. Increase lot size - I would have to trade 6-7 a day inherent Should it doubles.
    3. Change your earnings requirements/the initial half of the surroundings that you control (Save for a bigger balance).

    These are the 3 options at any time in this process if it appears your earnings requirement won't be able to be fulfilled. For many traders, particularly newer ones, the third alternative is always your very best bet.

    In case you are aware of how to properly handle your risk, I suggest choosing the second alternative.

    IMPORTANT NOTE: This shouldn't be attempted by new traders. I traded unprofitably for 4 years before I was able to eventually become profitable. Only once I became more profitable, was I could approach the markets like this, and hope that my numbers were right, and everything would work out at the ending ( a.k.a. on a large number of occurrences, I'd get my expected outcome. https://www.forexforum.co.za/general...oons-pics.html)

  7. #27
    Finite Infinite Games

    So now my maids arrived to clean my house office, which meant I had a few hours to do some reading. The book is called Finite and Infinite Games: A vision of existence.



    Only flipping through the pages, it is possible to find so many passages which use to both trading, and lifestyle generally. I wanted to mark down the passages I looked in now which I found really intriguing and why before I forget (which could be by tonight when I didn't post this lol).

    Below are just a few I found really remarkable. I am going to explain why for a few and not for others. I tried to list them in the best way I could to really show the train of thought.

    1. There are at least two kinds of matches. One could be called finite; another infinite. A game is played for the purpose of winning, an infinite game is with the intention of play.

    We play with a finite game, while Mr. Market plays an infinite one.

    2. An infinite game can't be played inside a finite game, although games can be played within an infinite game. Infinite players regard losses and their wins in whatever finite games that they play moments although in play.

    3. As it is essential for a game that is finite to really have a definitive ending, it must have a start. Therefore we can speak of finite games with temporal boundaries -- to which all players need to concur.

    Believe X-Axis.

    4. But players need to agree to the institution of numerical and spatial bounders . That is, the game must be played within a marked area.

    Believe Y-Axis.

    5. The rules will be different for each game that is finite. By understanding the principles, we know what the game is. What the principles set is a range of constraints on the participant... In the narrowest sense principles are not laws; they do not mandate specifical behavior, but only control the liberty of the players, permitting considerable space for choice within these restraints. -- Will go back to this.

    6. The rules have to be published before perform

    7. As they are set rules are not legitimate. They are legitimate because they are followed by the players of all games.

    8. If the principles of a finite game are special to that game, it is evident that the rules may not alter in the course of drama -- else there is a game being played.

    Think shing probabilities and rolling windows.

    9. It's on this point we loe the most crucial distinction between finite and infinite drama: The principles of an infinite game MUST change in the course of play.

    Again... shing probabilities, rolling windows. Is finite. The game Mr. Market plays is infinite.

    10. The principles of an infinite game are changed to stop anybody and to bring as many men as possible.

    Mr. Market.

    11. Because of this, an infinite game's principles have a different standing from those of a finite game. They are similar to a living language, where those of finite games such as principles of a debate's grammar.

    12. Finite players play within boundaries.

    Think of quote #5.

    13. Even though it could be evident enough that whoever plays a match performs it's frequently true that finite players come to believe that whatever they do , they MUST and will probably probably be unaware of the absolute liberty do.

    Think plogical trading. Chasing markets. Breaking rules, etc..

    14. Fields of play just do not impose themselves on us. Therefore finite play's constraints are self-limitations.

    15. So it is with all functions. Only can step to the role of mother. Individuals who assume this role, nevertheless, must suspend their freedom in order to function as the role requires.

    16. We shall refer to finite play. Script and plot do not seem to be written in advance, yet we are always able to look back in the path followed to success and say of those winners they knew how to behave and what to say. We shall refer to infinite players as dramatic, maintaining the future open, avoiding any outcome at all, and creating all scripts useless.

    17. Dramatically one chooses to be a mother; theatrically, one takes on the role of mother.

    18. Because the outcome is yet understood during the game, all finite play is dramatic. That the outcome is not understood, makes it a genuine game. The theatricality of finite play has to do with the reality that there's an outcome.

    19. The fact that there is a game that is finite dramatic means that it is the aim of each player by making a preferred end inevitable to eliminate its play.

    Consider how this can be achieved using number of occurrences and expected value.

    20. It's the need of all players that are finite to be Master Players, to be so skilled in their perform that nothing can surprise themperfectly edued that each movement in the game is foreseen in the start.

    That is exactly what people wish to be reality,

    21. A Master Player performs as though the game is already according to a script whose each detail is known before the drama itself.

    That is the reality. Think shing probabilities, number of occurrences, expected outcomes, central limit theorem, setting principles of postsecondary play...

  8. #28
    Is the thread based on your specialized trading and algos or is this also coinciding with yourfundamental analysis?

  9. #29
    Quote Originally Posted by ;
    Is your thread only based on your specialized trading and algos or is this also coinciding with yourfundamental analysis?
    This is only technical stuff and a few risk management material.

  10. #30
    Whites Reality Check

    Ever get the idea for a new system, backtest it and end up with a result like this?



    Hooray! It seems as if your new system is killing it! If these results can be trusted by you though how do you know for sure? Imagine if it was only a fluke?

    This is something which you want to figure out before investing in this new system with live money. And here is how:

    Among the of my favourite pages of Evidence-Based Technical Analysis is page 241. It's talking about Whites' Reality Check.



    So after doing all that, how does our completely new system stack up?



    Actually pretty darn good!

    In blue are the resamples yields and in red is the actual return of the machine.

    This system provides us a p-value of less than 0.01, meaning whether that system's rules expected return was truly equal to zero (see step 1) about 9 out of 10 times, the principles would make a return of roughly 17 percent or higher.

    The White's Reality Test is not really intuitive, but it's mathematically sound, which is all we care about.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
This website uses cookies
We use cookies to store session information to facilitate remembering your login information, to allow you to save website preferences, to personalise content and ads, to provide social media features and to analyse our traffic. We also share information about your use of our site with our social media, advertising and analytics partners.