Portfolio Doing Well

The portfolio I developed using StrategyQuant is working very well. I started it in a live account April 1, 2015 and it is up 14% and over 1000 pips.  Drawdown is minimal at 11%, my goal is to have drawdown capped at 20%, but ideally below 15%.  Although it’s too early to conclude success after only 2 months, the strategies’ drawdowns and smooth upward performance results are indicative of the back testing and forward testing results I had.  So it’s looking real good overall.

After I made 10%, I doubled the lot size.  This is a test live account (demo account was giving different trade results than the live account so I do not want to rely on a demo account’s results).  I am using a $1,000 account with a .01 micro lot to start, and am now using a .02 micro lot, and will continue to increase as profits grow to compound the returns.

The attached pic is of when I started this portfolio in a live test account April 1, 2015.  So it has accurate readings related only to this portfolio and filters out previous results from other strategies.  I also have the entire account tracked on MYfxbook (http://www.myfxbook.com/members/PrincipAnalysis/principle-analysis-sq-portfolio/1244436) but this documents the account since its inception which distorts the data related to this portfolio since I tested a grid trading strategy before and it was very wild so I stopped using it.  You can see on the MYfxbook link that the performance became significantly smoothed out once I started using the current StrategyQuant portfolio April 1st though.

I just wanted to drop by and post my results and at least short term success in using StrategyQuant to develop my own EAs.  So far so good.  I will be back again later with an updated report.

So, has anyone obtained profit already from her/his Portfolio?

WysÅ‚ane z mojego Z30 przy uÃ…¼yciu Tapatalka

It’s early days for me and I’m up a little, but no doubt there will be numerous fluctuations along the way…

I’ve been running a portfolio since Oct 8th with Oanda, and that’s currently up 21%, but I’ve had to endure a Max DD of 33% along the way…but no big deal.

I’m consistently refining the portfolio as my analysis of workable trading systems has improved (I kinda rushed to production, so to speak…) and since November 10th I opened another account with Global Prime with the more refined portfolio (which is of course constantly being refined) and that is currently up over 15% with Max DD at 15%.

It’s somewhat remarkable (although somewhat expected as well) just how different the portfolios run with different brokers with the different spreads and execution etc…I can be in and out of a trade with GP inside a day whilst the same trade executed with Oanda can still be open days later. Anyway…just a point on how important a good broker is which is something I have probably never really appreciated until now. That said, profitable strategies can be created for most brokers, it all depends on your data, testing methods, and analysis.

Back to the point though…yes I’m in profit but it’s early days.

As I mentioned in a previous thread Brokers will also have an impact on the strategy results and @RJL mentions above

I have exactly the same portfolio of strategies/risk settings running on two different brokers on live accounts (different balances though) same starting times, have only been running since end of September.   These portfolios are targeting a Max DD of 10% and annual return better than current Australian bank interest rates of 3% annually.

GoMarkets Australia and Pepperstone Australia, both are pro/paid type brokerage accounts.

goMarkets Pro  YTD +2.91% 

Pepperstone Razor  YTD -4.22%

So if you’re finding that the strategy was robust etc in backtests but not in live trading try a different broker.

I also think that its important to monitor each strategy in the portfolio using SQ Analyser and MC analysis, for each strategy remove or reduce risk settings if getting close to Worst case drawdown DD, because you don’t need many strategies in a portfolio to start having a run of losses before your portfolio is hitting your WC DD especially if using %based MM.  

Remember your portfolio is managing your hard earned capital, your job is to maintain and grow it, not loose it!  If you don’t adhere to your MonteCarlo WC DD then you are just asking for trouble. 

Unfortunately Automated portfolio trading is not a set and forget way of trading (as much as I’d love it to be) it constantly needs to be monitored. 

Seaton: the choice of broker makes no difference if you are not HEAVY scalping on the 1M or something, the strategies still go down overall, each time – everyone can “sing a song” about this here. I am doing this since 8 years now, nothing is really working out, not even with regular re-optimizations / adding new strategies. As much as I hope it for you, your portfolio will most likely be no different – I´ve seen it literally a 1000 times in my life. Apart from the small differences between your brokers, I can almost guarantee you where it will go in the mid to long run. But why not proof us different and show us that you beat the 3% bank rate? Give us a MyFxBook link to your portfolio 🙂 I seriously hope it will turn out for you, but the chance is as low as winning a lottery. Otherwise I´d made have millions already if all my (robustness and boot-strap tested!!) strategies would ever have made me money longer than a few days or weeks at all.

My conclusion is simple and logic: the markets are changing to drastic, especially nowadays. Everyone is switching to automated trading and the institutionals are switching to HFT, so the underlying inefficiencies that our strategies explore, are gone by the day we go live as the market keeps on transforming. Really, I´d love if it would be different, but show me just ONE mid / long term successful EA (apart from SQ) – especially in 2015. There are forums all over the internet that hold thousands of EAs that have been made throughout the last 10 years when forex forums became popular, yet NOT ONE of them is making money constantly. The same here, the strategies section of this Forum is as empty as it can be and the “successful” portfolios that HAVE been posted here, are now either deleted or, if still available on MyFxBook, go down or max side-ways after a slight initial gain.

Doesn´t all of that tell us something? I am sorry, I am not pessimistic usually, but I´ve spent 8 years on all of this, in heaps of different programs (MT4, Multicharts, SQ, PriceActionLab, TSL, Kantu, FIX API, you name it), have learned tons of different languages, do log my own tick-data of my broker (which has great execution without to much slippage and I actually kind of work for them as well as an adviser for liquidity), yet nothing has worked longer than a few weeks at all. Really, I´d love to be convinced differently, so everyone who is successful with automated trading, show us your MyFxBooks, or at least a screenshot of a successful LIVE account (you can remove account number etc) that is traded automatically (regardless if SQ). I have yet to see just ONE which does more than a few weeks at all before failing. As hard as it is….

I´ve collect all the MyFxBooks links that I was able to find on the Forum that do run SQ generated strategies and that have (had) a solid approach to how they did it, see for yourself where all of them are (or have been, as they have been stopped) going:

http://www.myfxbook.com/members/strategyquant/portfolio-1/1188718

http://www.myfxbook.com/members/Threshold/thresho1d-portfolio/1238851

http://www.myfxbook.com/members/HarmonyTrade/harmony-trade/1390266

http://www.myfxbook.com/members/PrincipAnalysis/principle-analysis-sq-portfolio/1244436

http://www.myfxbook.com/members/adrian8891/eurusd-short-test-incubation/1173012

http://www.myfxbook.com/members/adrian8891/portfolio-gbpusd/1123911

http://www.myfxbook.com/members/adrian8891/runner8/1095185

What to say about that? Alone the low amount of MyFxBook links shows that most must not be so proud to show their results, yet the ones that do (thanks for that), are all going nowhere or down after the typical initial growth of a week or a month at the most. Sorry to say, but it´s almost time for me to call this done, it´s been repeating EXACTLY like that for 8 years for me. And I see it´s the same for anyone else here as, beside advising theoretical replies (thanks for that though) why “it must be done this or that way instead” or why “it is clear that it is failing and you just must do X”, which I am hearing since 8 years as well, I have yet to see just ONE person posting their successful, automated, FX trading account (regardless of the method the system has been created with).

I did a quick search of the Internet for any good commercial EA’s, based on real verified live trading.

This is one of the better ones:

http://www.myfxbook.com/members/fxrobotreviews/vortex-trader-pro/1148041

As you can see 2015 looks very poor.  Same for some other I’ve looked into (to gain ideas on trading strategies).

My own results are similar, this year has been unkind.  I think there’s a lot going on in the global economy, and it’s chaotic.  We are probably on the verge of another economic/financial crash like we witnessed in 2007/2008.  I such circumstances I am happy with any system that doesn’t lose money.  Better times will come.

PS Here’ another commercial EA that is having a poor time recently.

http://www.myfxbook.com/members/channeltraderpro/channel-trader-pro/1283833

This is investment fund that I manage, I dont use strategies from SQ, but so far results are good enough for me https://www.mql5.com/en/signals/136822

WysÅ‚ane z mojego Z30 przy uÃ…¼yciu Tapatalka

Seaton: the choice of broker makes no difference if you are not HEAVY scalping on the 1M or something, the strategies still go down overall, each time – everyone can “sing a song” about this here. I am doing this since 8 years now, nothing is really working out, not even with regular re-optimizations / adding new strategies. As much as I hope it for you, your portfolio will most likely be no different – I´ve seen it literally a 1000 times in my life. Apart from the small differences between your brokers, I can almost guarantee you where it will go in the mid to long run. But why not proof us different and show us that you beat the 3% bank rate? Give us a MyFxBook link to your portfolio 🙂 I seriously hope it will turn out for you, but the chance is as low as winning a lottery. Otherwise I´d made have millions already if all my (robustness and boot-strap tested!!) strategies would ever have made me money longer than a few days or weeks at all.

 

As I mentioned above it is only early days on my accounts so imho is too early to tell, but I am seeing a marked difference between both accounts, running same portfolio/risk settings from the same VM, but with different results, and but will continue trading until portfolio WC are hit.  I’ve already posted the link earlier.

That said, I myself tend to agree with all your points and only time will confirm for me.  My thoughts are that there has been a paradigm shift with mechanical trading on all markets not just forex as Quant and Machine learning is so accessible to the masses, so my thoughts are that the core underlying principle of Quant trading and ML i.e. backtesting on historical data, is now irrelevant due to this paradigm shift and the change it is causing in more efficient markets and why institutions have moved to HFT (just as you have said), but again time will tell, I guess like most a bit of me is still optimistic it can be done 🙂

However in saying that, there are still trends, so until then I will still go about my business trying to find and exploit some rather small market inefficiencies until I’m 100% sure that this is entirely the case.

Like yourself I also like to keep track of fellow SQ users to gauge progress, unfortunately we’re not privy to the modelling parameters for the strategies and portfolios, but is my guess some have fairly high risk associated with the DD I see on these and with the extended DD period most seem to be experiencing then I’m not surprised at their results.

Aside from this, I’m also a member of Asirikuy.com and they have approx 1600 mined strategies available to all members.  the only downside their platform is not the easiest to get up and running or use, which is why I love SQ.     A number of these are tested on live accounts and showing very good success, although infrequent trades.   A method they employ in their mining is testing DMB (Data Mining Bias) of the strategy,  This effectively tests for randomness, The higher the DMB the more likely the results are no better than random, those with low DMB results are more than likely a good strategy, remember all based on historical data so how valid going forward is this? again time will tell.   I would like to see some form of this in future versions of SQ.  Unfortunately I can’t share the links to the Asirikuy MyFxBook accounts as is against membership policy, but I monitor these regularly.

Daniel does run a public blog http://www.mechanicalforex.comif you’re not already familiar with it. 

Heres Link to his post on DMB

http://mechanicalforex.com/2015/06/data-mining-bias-a-mathematical-look-part-1.html 

Stephen 

@geektrader I forgot to add:

With the accounts I monitor as do you, I would ask myself how some of the portfolios are managed?   As some of the Drawdowns / Risk I see   i.e going from 0 equity to near 100% equity growth then back below 0%

To me some of these are not being managed properly or the risk settings are set way too high.  Remember in portfolio theory a portfolio of strategies will reach WC DD quicker with individual strategies still set at their normal risk settings.   The more strategies in the portfolio, the lower the risk of each individual strategy should be.

The questions I would ask myself is what strategies are failing in the portfolios of the links and why haven’t they be taken offline and replaced with new performing ones?   Again to my points above, I’m not privy to the design parameters and what the worst case drawdowns are for each strategy or their risk levels, maybe they are performing to within their design parameters, but my guess is the strategies have been added, perform well for a bit then are failing, but not taken offline, or have risk reduced, hence bad performing portfolio results of some.  My aim is to protect my capital, man, its too hard to come by easily 🙂   unfortunately trading is not as easy as one would hope sigh!!  definitely no retirement for me yet.

Others I see look ok but in a DD period, again unless we can see the SQ analysis of these strategies we have no idea what state there are in and we are only guessing.

Maybe this is the problem because of more efficient market is profitable Strategies are now only profitable for a shorter time where historically they lasted longer without tuning them or taking offline?  Or maybe the market in general is in draw down and will see recovery at a future point who knows?

This leads on to a post I wrote a while back to open up discussion in the feature request section of the forum about a trade manager, where it takes the hard work out of the above management of failing trading systems, but also enhances profitable ones, where the trade manager will shadow trade strategies and adjust risk on actual orders based on MC analysis and previous trades and if WC DD is hit will not place order until shadow trades are back to normal.  Or will scale up or down Risk of orders depending on performance.  This is based on the work of Howard Bandy.

Stephen…

Hi Stephen,

thanks for the good and grounded reply, I much appreciate it.  Personally I don´t really look for equity, but for the pips equity curve on these accounts and as you say – so money management in terms of lot-sizing, while important, is ignored by me for that fact. But let´s leave away the failing strategies that we see everywhere…. one interesting find I wanted to report in that relation is the tick-data analysis (Dukascopy from 2007 to now and from my own broker from 2015 January to now) I´ve done this weekend.

I am running a sort of “HFT” scalper as you could call it since a few years. It´s kinda interesting that exactly since end of December 2014, especially EURUSD has had a complete change on the 3 seconds time-frame (this is where this scalper works). The entry logic it is using nowadays and that works good now, did NOT EVEN EXIST just even ONE time in the EURUSD tick-data prior to December 2014. But from December 2014 (around mid) and to date, it had ~10000 trades on that new entry logic on the 3 seconds timeframe.

What it is using and what I also verified by eye could be called micro-spikes that are really fast (1 to 3 seconds long, 8 to 10 pips in size) – those are completely new since December 2014, they didn´t exist before, EVER. Now they are going on in the tick-data ALL the time since then and also on different brokers, so I can exclude that this is just a broker dependent behavior. They got even more heavy since June 2015 and have kept on going like that. Isn´t it interesting that this is exactly the time since when most of my strategies (and most other EAs like the ones posted above) have started to fail? In any case, something is going on in the market since December 2014 and especially July 2015 that never existed like that before. What makes me wonder though, is that it right away start in mid of December 2014. Sure it´s getting heavier since July 2015 (and this is the time since really all my strategies are going side-ways or down), but it is like someone has “switched on a computer algo” on that day in mid December 2014 that since has changed the markets.

I will try to investigate a bit since I have access to a NY4 liquidity aggregator and can see the order book, but since Forex is OTC, it would be hard to find who is triggering that. Maybe I should actually even report those findings to the SEC or NFA, I´ll see.

In any case, it IS there and it HAS changed the market quite a lot. For me, that´s the clear reason things are failing “out of sudden” since that time. The only thing we can do now is to generate new strategies that also can handle the current market behavior. The more of that is happening and the longer it is happening, the more we can start to mine the inefficiency it is using and adapt our strategies to it. If my “HFT” scalper can explore the inefficiency and ride on it, mid term strategies should be able to do so too, just a question of time until SQ will have enough data to generate strategies on that new behavior. And should everyone switch to automated trading, there will be even more exploitable patterns since we all know, computers can only be pseudo-random, hence always follow a specific algorithm when doing their trades. Only humans can create random trades. If they are eliminated, previous patterns my fail at first as everything is shifting to automated trading, but once this move is complete, there will be more patterns than ever before thanks to that everything is algorithmic traded and hence predictable as it never can be really random like a human can be.

As for Daniel, yes, I know him, in fact I´ve been fixing bugs with him on OpenKantu the last days. If he had a monthly subscription, I´d have joined already, but paying a year in advance is honestly not something I want to do without seeing results, which he doesn´t want to give out. His software might be complicated, but sounds great by the approaches it uses and surely would be something for me (I am coming from Linux). Anyhow, Zorro (some people here might know it), sounds great and even better too, yet no one is making money with it (see their Forum) and it actually has everything and even more than what Daniels software has. Yet they fail… and yet Daniel does not want to show any live results. And in the end, that´s the only thing that really counts, despite all theory of what SHOULD or COULD work theoretically (and most often simply does NOT).

Well, personally I don´t know where my journey will take me from here, maybe I stay in Forex, maybe I am out finally. But I wish everyone best of success at the least:)

Geektrader omg man 😉 Take some time off and think about what you have found so far. Works only for weeks means WORKS, you just have to develop it. Don’t you realise that you are ACTUALLY predicting (some) future and in not so easy to predict area? 😉
Take a look at this http://www.myfxbook.com/members/matka/three/32338/sehi6XmNT9FPHLVN2iS9
Worked just for a week
cheerio

That account is from 2010, trading privileges are not verified, broker is not named and is most likely latency arbitrage by the gains it has and the drawdown it got as well as the trade duration (0 seconds) and the average win and loss. What should this demonstrate in relation to SQ? SQ did not even exist in 2010. And with latency arbitrage it´s no problem to make gains like that, see:

That´s also the reason why your account only worked for one week and the one above will most likely be stopped soon too, since brokers notice this and terminate the account + profits and hardly allow you withdraw profits gained this way. It´s fun to do latency arbitrage, but in the long run it´s hard to get any kind of withdrawal from a broker that you cheat like that, they all usually notice it and simply cancel your profits and kick you out.

However, this all has nothing to do with real automated trading by exploiting historical patterns, so your post is pretty useless, sorry. The “works for a week” in relation to SQ was more of a phrase than reality, most of the time even the first week is negative, especially in 2015 (read previous posts).

I think the main reason why SQ strategies are useless for you is in your mind. You actually wrote how to use in one of your posts but you refuse to do it.

Please don’t call me cheat, it was not me who manipulated quotes and I did not break single paragraph in any agreement.

Im sorry if my post is useless for you.

No, not in my mind, but in 8 years of doing this it has proven to get nowhere in the end – and that´s the same for most users who´ve messaged me the last days in private saying “I am in the same situation”. For some reason people just don´t like to admit they´ve failed in public. But honestly, after 8 years of wasted efforts with almost 10+ hours each day doing nothing else than developing automated systems in heaps of different platforms and methods throughout the years, I don´t care anymore to just being honest.

Anyhow, why talk so much, show us your profitable SQ strategy or portfolio of strategies. I´d really really love to be proven different, but I´ve read phrases like yours ever SO often on all the forums… that it is “just the next update to the platform and we can then do this and that and can then finally be profitable” (through a ever higher amount of curve matching because even more parameters can be optimized), that “it is just the current market, once it “turns”, it will all work again”, that “the portfolio is just in a typical drawdown phase right now, no way to avoid it and long term it will be profitable”….. you can add your phrase here. Honestly, I read this on each and every Forum, independent from the platform that is being used since 2006 – be it MultiCharts, Adaptrade, 1000´s of MT4 EAs, StrategyQuant, Price Action Lab, whatever…. and it´s just tiring. No one can show any good longer term live results, all the EAs on Forums you´ll find (just see for example Steve Hopwoods Forum, a complete no-bullshit one, yet, after having coded hundreds of EAs and ideas, NONE of them was nor is profitable). The best you see is some side-ways action, most often it simply is a straight and ongoing drawdown. It´s just the reality unfortunately.

But then again, let´s show some profitable longer (or even mid term, say half a year) that really go UP steadily trading LIVE – I haven´t found one here, neither on any other Forum I visit. I´d love to see it! But then again, please don´t waste our time with a latency arbitraged, unverified broker account from 2010. And if you say you haven´t breached anything ever: then what was it that made the X-thousand % there? Believe me, I smell latency arbitrage from a mile, because I am doing it as well – and the average trade length, average winner and loser size, amount of trades, pip expectancy and especially % gain in relation to drawdown and in relation to the time the account was traded and especially that it stopped trading after just a week (got busted, right?), can be nothing else than latency arbitrage. NOTHING else takes a initial deposit of $500 to $420000 in a week. But if I am wrong, why not tell us what it was and why don´t you still trade it if it was so great? If you made 420k that week in 2010, I am sure you made another 420k the next week and would be millionaire right now without any need to hang out in a Forum like here and without the need of trying to create strategies in SQ – because then you´d still be using that strategy – or what am I missing? Truly sorry for the harsh words, but that´s the typical bullcrap you come around in other Forums as well, which does nothing else than to make newbies think that the “big buck” can be made in Forex through automated trading. That´s far from the reality for 99% of these traders that go home bankrupt 2 months later and is just plain misleading them.

I showed this account as an example, hoping that something will click for you. I am sorry if someone took it as an offence.
With all respect, I don’t understand why you demand my verified personal ratings, I am not here to proove anything and this is not a good place to publish finance related info.
However you are right saying that it is not possible to build simple rules, mt4’s like strategy and hope it will work forever. Simple maths and logic tells that. But it is just another and one of many problems we have to solve.

Sorry mate, but a latency arbitrage account show doesn´t help anyone here.

It´s also the typical answer when asking such questions “don´t want to share my financial informations”. Why not simply admit you don´t have anything successful to share either? And you don´t need to share any financial informations btw, just make a MyFxBook link and hide all the facts like € $ amounts, just leave the percentage turned on. That´s all I am saying… yet, I´ve never seen such a link anywhere in 8 years:)