BlueHorseshoe

You're just starting out and your main concern is tax implications . . . seriously? If I were you I would be more concerned about how I could offset my losses as a tax cost . . . That sounds terribly unhelpful, but that's the reality as a 'beginner' (your words not mine). Bluehorseshoe ps. Supposing you're actually going to make a decent return, here's a rough blueprint: - Set up as a business in an offshore jurisdiction. - Allow profits to accumulate and compound them within your strategy (yes, that means you don't get to spend any of the money you make at the time you make it). - Once you've accumulated a decent amount of money move to that jursidiction, extract and enjoy the profits. What, you don't want to live in the Bahamas? Tough luck . . . - Alternatively, extract the profits from that jurisdiction and pay the tax within your own. Through compounding you'll still have made far more in real dollar terms than by removing profits on a regular basis. It's your choice.

I have only met and observed one successful discretionary trader. She was successful from the outset, and she was unable to tell me anything whatsoever about how she traded, beyond 'I buy it if I think it's going to go up'. She knew nothing about technical analysis and didn't even watch the DOM - just watched the price whilst doing her (rather well paid dayjob) and made a few trades a day, with a very high percentage win rate and very small losses. For reasons that I am not prepared to go into here I know that she had done this for a sustained period. This was obviously someone with a 'gift' (or whatever you want to call it) for understanding price action. If you don't have this innate ability, I think that although it may be possible to develop it to a lesser degree, a trader is essentially better off not trying - at least by embracing a mechanical approach the trader can have a high degree of historical confidence, even if that confidence is not justifiable going forward. I fully accept that you and others may disagree with me. If so, then tell the thread author not me, as it's his future in the making (I am rubbish beyond belief at discretionary trading and abandoned it a long time ago . . . Although I occassionally wonder about opening a very expensive online trading room where I make terrible discretionary predictions and then traders get rich by fading my calls!). Best wishes, Bluehorseshoe

One of the primary divisions within this forum (or within pretty much any group of traders) is that between discretionary and mechanical traders. Both can be successful . . . And both can be disastrous. My advice (which others may disagree with) would be to try discretionary trading with a sim account. If you're not obviously successful at this within a few weeks then you probably don't have the knack for discretionary trading and should move on. Successful discretionary traders are an exception to the rule, and like all exceptions to a rule, there are exceptions to the exception - hence there are one or two on this forum who developed discretionary trading skills over a prolonged period from more mediocre initial results. If you don't have 'the knack', you will need to consider whether you want to devote a hell of a lot of screen time trying to develop it (a goal you may not suceed in achieving). Should you decide to go down a mechanical road, then there are countless pitfalls that await you there as well. Unfortuantely pure luck is one of them, however much many refuse to acknowledge this. Bluehorseshoe

Optimised Fixed Dollar MAE Derived Standard Deviation-based fixed/trailing ATR/Simple Range-based fixed/trailing EL 'Volatility'-based fixed/trailing Parabolic Moving Average 'Chandelier' Hanging Stop Donchian N-period high/low Channel Or for something truly adaptive: Sell longs when a 2-period RSI crosses above 75 and cover shorts when it falls below 25. Hope that gives you plenty to think about. Bluehorseshoe

Art Collins 'Beating the Financial Markets' - a recommendation from someone on this forum, in fact! I like the authors simple and upfront approach, but am struggling to relate to many of the strategies, many of which are momentum based and have not held up very well in the time since the book was published. Also, I am about to read an interview with James Simmons that I just printed off - hopefully it will help me get to sleep! Bluehorseshoe

. . . You start arranging food on your plate to look like candlestick formations Bluehorseshoe

I learnt to program in BASIC many, many years ago, when I was about eleven! I was wondering whether clubbing together a series of 'FOR N=A TO B . . . NEXT' type statements would work? If so, could anyone provide a bit example code to illustrate how this would work in EL? Also, I have clearly confused matters by my use of the word 'String'. I am just using the term as a synonym of 'Sequence', and not in whatever special sense the term functions within EL. Although I am only just begining to explore this whole concept, I doubt that I will want to follow the market on anything like a tick-by-tick basis. The purpose of the code would be to draw out sequences that have demonstrated high predictive properties within a historical data set. Thanks, Bluehorseshoe

Hi ZDO, Thanks for your reply. Sorry my question isn't as clear as it might be - I think to convey exactly what I mean would require an explanation of the full context, which could take pages. But I'll have another go: Suppose that I wanted to know whether, following three down closes, an up close or a down close was the more prominent within a historical data set. I could define the series/string/sequence(whatever it should be called) 'DDD' and also create two variables 'DDDD(0)' and 'DDDU(0)' each of which has a "predictive" value for the fourth day. For each instance within the historical data where the sequence 'DDD' occurs, the variable 'DDDD' would predict a fourth down day, and 'DDDU' would predict a fourth up day. By examining the actual outcome on the fourth day, I could then "reward" whichever of the two variables correctly predicted the actual outcome by increasing its value, and "punish" whichever of the variables predicted it incorrectly by decreasing its value. So, in the case of a sequence of closes DDD then the following would happen: If c<c[1] and c[1]<c[2] and c[2]<c[3] then begin If DDDU>DDDD then P=DDDU Else P=DDDD; End; End; If P[1]=DDDD and c<c[1] then begin DDDD=DDDD[1]+1; DDDU=DDDU[1]-1; End; If P[1]=DDDU and c>c[1] then begin DDDU=DDDU[1]+1; DDDD=DDDD[1]-1; End; Over time, if three down closes were suceeded by an up close more often than not, then the variable 'DDDU' would grow while the variable 'DDDD' would diminish. The variable 'P', by the way, is part of a slightly different concept of giving precedence to the sequence which has shown the greatest historical predictive capability. In the example above, I have chosen to examine historical probabilities surrounding a particular sequence (three down closes) which I have manually specified. What I am looking for is a mechanical way of processing all possible sequences of a given length in the same fashion. Hopefully my aim is now a little clearer. Thanks. Bluehorseshoe

In my opinion, the main thing that you need to know about the @ES is that it is heavily 'mean reverting'. This means that it spends most of its time backing and filling, retracing earlier movements and chopping around, never really going anywhere much. It's average daily change is miniscule. In the simplest terms, wherever it is headed right now, it will soon be coming back the other way. Trending, breakout exceptions do occur, but they are few and far between, and often they are just a reversion to a longer-term mean in a higher timeframe. In my opinion (which may not be other's), your approach needs to be based around trying to sell strength and buy weakness. If you get good at this, then you will need to try and find ways to make the more advanced distinctions that MightyMouse describes. As you are already familiar with the RSI, here is an example of the type of approach you could take: - If the market is in an uptrend then buy on the close whenever a 2-period RSI falls below 5. - Exit your long position when the 2-period RSI crosses back above 60 (whether for a profit or a loss). You'll need to find a way of determining the trend. A simple MA might do until you find something better. What was the best length (historically) for the MA? And why use a 2-period RSI rather than a 5-period RSI? And why use 5 as the oversold level? To answer questions of this sort you would need to learn to program and backtest your strategies. I would imagine you would want to use something like an 80 period SMA. You might also want to consider the trend in higher timeframes and align yourself to this. You could only take longs from a fifteen minute chart when both the hourly and the daily chart were also above their MAs, for example. You might want to try and refine your entries as well. Instead of entering on the close when the RSI falls below five, wait for this to happen on your fifteen minute chart and then wait for it to happen on a one minute chart to give your actual entry. Does this work any better? Once again, you'd need to learn how to mechanically backtest this in order to get anything like a reasonable answer. I hope all that helps you and you can find ways to build upon it. Bluehorseshoe

I would take this a step further: if you don't understand how each is calculated and what that calculation is intended to achieve, then you won't know how to use the indicator. A completely ridiculous sounding (but true) analogy is this: Each day I drive my boat to work on the road. One day my boat won't work. I don't understand anything about my boat or its engine or the processes that go on inside it, so I take the boat to a boat mechanic. The mechanic has a look and says 'well there's no wonder it's not working, mate, you're trying to drive it on tarmac roads - this vehicle is meant to go on the water!'. I reply 'but I've been driving the boat to work on the road every day for three months now', to which the boat mechanic shrugs 'well you got away pretty lucky not having a very nasty accident pal: this is a boat, and it just ain't designed to go on tarmac. . .' We're used to taking things for granted. In reality I drive a car to work on the road everyday because everybody tells me that this is what cars are for. But I know nothing about cars, so ultimately I'm just making a big assumption. This kind of assumption tends tends to work well in real life and it would be pretty difficult to function without it. In trading you can't afford to make these kinds of assumptions; you need to know exactly what's going on under the hood. This does not mean that you need to know the formula for every indicator under the sun, however, because you won't need to use every indicator under the sun. I hope that's helpful.

Hi Onesmith, Thanks for your reply. Unless I'm missing something, then with either approach I would still need to manually set up each possible permutation. This is easy enough in the example I gave where there are only eight permutations, but supposing that I had five possible values ('upstrong','upweak','flat','downstrong','downweak') and considered strings of up to ten values (ie periods) in length, then the number of permutations would be far too large to type. This would pressumably be possible by using the optimisation feature with a strategy report, in which every possible permutation of inputs would be tested. Is there a way to do this on an ongoing basis internally to the code? Cheers, Bluehorseshoe

Hi folks, Wondering if anyone can help me . . . I'm playing around with a genetic optimisation concept, and I'm wondering what would be the best way to get EL to return each individual permutation for a set of values. Obviously I can easily calculate the n! factorial, but my requirement is for the code to return (and then process) each individual possible permutation as a string (I use that last word loosely!). Oh, and just to make it a bit more complicated, I want to include repetitions of values, up to and including strings composed entirely of a single value. Hopefully I can make this clearer with an example. Considering the last three periods, and whether they were 'up closes' ('U') or 'down closes' ('D'), the following permutations would be possible: UUU UUD UDU UDD DUU DUD DDU DDD So, having defined the values 'U' and 'D', how could I instruct EL to return each of these 3-value strings sequentially? Thanks Bluehorseshoe

I agree with all but your very first statement. A statician will not tell you that 30 data events are significant; they will tell you that 30 data events are significant for certain types of data. The exotic data produced by the financial markets not does not fall into this category. Other than the above, I am in total agreement with you. My original post was just pointing out, regardless, that the strategy does hold up well even across greater sample sizes than the 60 trades the thread starter shows.

I can't suggest a way around the problem of being unable to make a paper-based application whilst in Afghanistan, but I believe that a funded Tradestation Forex account is free in perpetuity with no minumum trading requirement. The platform is identical to that for futures, stocks etc, and you get charting for pretty much every stock under the sun, a selection of futures contracts including the @ES, spot forex (obviously), and access to the EasyLanguage development environment which you can use to backtest your trading ideas. If possible, try giving tradestation a call and chatting to one of their sales reps - it's not that hard to twist their arm into getting them to offer you free trials etc as long as you come across as someone who will generate them commissions down the line I hope that's helpful to you.

Is the omission of 'sell high buy low' from your statement intentional? Bluehorseshoe

The profit or loss on my account is the most important indicator, but it's not always my favourite!

The flash crash would actually be a very good example of why not to have a stop in place (in the case of this specific strategy). The flash crash would barely have made a dent in your equity curve, as the maket bounced significantly the next day. So, assuming you could meet margin calls throughout the crash you would have been fine. Last summer's mammoth sell-off, on the other hand - well there's the time you needed a stop in place alongside the RSI!

You seem to have a bit of a rose-tinted view, but to some extent, isn't a forum like this one just a modern electronic 'community of gentlemen'? Admittedly there is the odd thief around, but I've certainly gained plenty from just a few months of talking with other traders on here. And if I have a specific query, I can get brief and to-the-point input from countless people, rather than relying on one particular mentor's viewpoint. Although none of this is perhaps the same as me going and sitting in front of a computer screen with a coach for a week, it certainly has other advantages in my opinion. Bluehorseshoe

While I fully appreciate the point that you're making about people's ability to muck up even with a profitable strategy due to their own psychological makeup . . . has it occurred to you that mistakes such as the one you describe could be the product of poor teach and a student left not having fully understood your strategy? Or a genuine ambiguity or subjectivity in the strategy? Bluehorseshoe

If you investigate the source (Connors) of the strategy that JSwanson is discussing in brief here, you will find extensive testing over a wide range of instruments over many decades. As the strategy concept is to sell strength and buy weakness, you can also find other ways to test the general concept without relying on the particular triggers given in this strategy (in fact, JSwanson shows the results of a 3-Up Close, 3-Down close rule in another thread). Of far more concern that the limited number of data points would be the number of variable parameters: length of MA, length of RSI, OB level of RSI, OS level of RSI, RSI input data (close as opposed to open, high, low, avgprice etc). You would want to see how robust the strategy remained as the parameters were varied. Hope that's helpful.

If you're especially keen to trade with leverage, then you'll probably find it easier to obtain by focussing on instruments that a) are highly liquid, b) don't gap (24 hour session, or Cash plus Globex sessions), c) cannot lock limit. This means focussing on instruments like stock index futures. As a daytrader, for example, you can readily control a single E-Mini S&P contract, which recently has had a value of between $60 and $70k, with about $500 in your brokerage account. Now I'm not suggesting that you should do that . . . but it gives you a good idea of where leverage is easy to come by. Hope that's helpful.

I would be careful with this approach if I were you. A lot of the strategies that I trade or track performed conspicuously well in that 2008 period (to the extent that the equity curve pretty much brickwalls). This is probably because they are contrarian in nature and therefore doing roughly the opposite of whatever your strategy is trying to do when it struggles in 2008. However, if I developed such a strategy now and then used 2008 for my out of sample period, then I would get a wildly optimistic and hopelessly wrong-headed impression of my strategy's likely performance. . . . A much more interesting period to look at is 2007. This is when the volatility really started to ramp up, as all the higher tranched credit default swaps etc went belly-up behind the scenes - how did your strategy respond to this steadily increasing volatility? Did it adapt? Hope that helps Bluehorseshoe

After you've typed your code out in the reply box, select it and then hit the ' # ' symbol on the kitchen sink above - this will put the code in a box in a courier style font, and keep Tams happy. Also, if you select your code and then use the symbol that looks like a flashing dollar sign, this will cause TL to forward your code to the guys at RenTech, who will give it a good look over and then send it back to you with suggestions for modifications and improvements, along with a complimentary example of some of their own strategies. Or maybe that was just a lovely dream that I had . . . Bluehorseshoe

Hello, In your other thread you have been discussing the importance of look-inside-bar resolution in backtesting. The following code should (hopefully!) be an adaptation of the above code for implementation on a lower timeframe (eg 5 min) chart, where it is assumed that you're wanting to trade data points from a daily chart (which I imagine was the intention of the original author as you say it has come from an old book and few people constructed intra-day strategies way back when): If (opend(0)>highd(1) or opend(0)<lowd(1)) then Buy next bar at closed(1)+(highd(1)-lowd(1)) stop; If (opend(0)<lowd(1) or opend(0)>highd(1)) then Sell next bar at closed(1)-(highd(1)-lowd(1)) stop; I haven't actually checked this, so hopefully it works! You might want to experiment with adding a condition that limits the number of trades to one per day. You may also want to give some thought to how the strategy should deal with an open that, for instance, is greater than the prior close+prior range, and could therefore trigger your stop way above the desired entry. Bluehorseshoe

Thanks for an interesting and useful reply, Steve. It's a while since I looked at activity around the close in any detail, so your response has re-ignited my interest in this. Cheers, Bluehorseshoe