Diversified Trend Trading Approach | Page 23 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Is Grid Trading (combined trend following and counter trend) Profitable? 3 replies

Making a good diversified portfolio 10 replies

Diversified EA Money Management Code 10 replies

HAMA PAD - A Simple Trading Approach 7 replies

Trading a (semi) automatic approach...but want more 11 replies

Attachments: Diversified Trend Trading Approach
Exit Attachments

Diversified Trend Trading Approach

  • #441
  • Sep 3, 2015 11:46pm Sep 3, 2015 11:46pm
Quoting AshR
{quote} Hi C, I'm a fast reader :nerd: Ok, so after about 5 pages, I skipped forward a bit to read the recap post I'm trying to develop my own trend-trading approach, and my one big take-away from your approach is the use of the SDC. Amazingly, a 2 SDC works better with my approach, since I try to stay in longer. Thanks so much for the deets on IG Markets. Oanda is Asic-regulated in Oz, and from my perspective comes out sparkling clean (-ish) from the CHF mess. I'll look into all that info about IG markets in a moment
Nice Ash.....Your thoughts regarding a 2 SDC are very appealing to me. I have been listening to a few podcasts from Top traders unplugged and in particular the ideas of a few of the notable trend traders (CTA's) who advise that since 2008 and with subsequent government intervention, the market has seen increased noise on the short term resulting in trends becoming more long term in nature and forcing the trend following ETA's to apply longer range techniques.
  • #442
  • Edited 10:03pm Sep 10, 2015 8:14pm | Edited 10:03pm
Quoting Copernicus
{quote} Nice Ash.....Your thoughts regarding a 2 SDC are very appealing to me. I have been listening to a few podcasts from Top traders unplugged and in particular the ideas of a few of the notable trend traders (CTA's) who advise that since 2008 and with subsequent government intervention, the market has seen increased noise on the short term resulting in trends becoming more long term in nature and forcing the trend following ETA's to apply longer range techniques.
Well, I'm a discretionary-method person, but sometimes (always?) it's hard to have faith in your discretion. I've found that applying a 2SDC channel and then choosing the swing high above the 2SD mark seems to work well for me in terms of SL (for a short trade), and it's easier to pick instead of having to choose from about 4 different recent swing highs.

Thanks for the podcasts' link! I'll go through a few of the transcripts soon, they sound like very interesting interviews. My own method involves holding for an aeon, until the fundamental drivers change (when I lose faith in my judgement of fundies, I can turn to COT charts to see if speculators' net positions have changed).

Good luck with your EA development. At a quick glance, I find that a lot depends on the length of SDC and the timing of the entry. With long term trades, there's often an initial trend, consolidation (during which the SDC might get hit) and then another drive higher/lower, and repeat.
  • #443
  • Edited 9:03pm Sep 13, 2015 6:12pm | Edited 9:03pm
Quoting AshR
{quote} Well, I'm a discretionary-method person, but sometimes (always?) it's hard to have faith in your discretion. I've found that applying a 2SDC channel and then choosing the swing high above the 2SD mark seems to work well for me in terms of SL (for a short trade), and it's easier to pick instead of having to choose from about 4 different recent swing highs. Thanks for the podcasts' link! I'll go through a few of the transcripts soon, they sound like very interesting interviews. My own method involves holding for an aeon, until the fundamental...
Hey Ash :-)

While plotting where the SDC commences may be a simple exercise using a discretionary approach, it really sucks when you try and get an EA to do it. We have attempted to overcome this hurdle by creating an EA that iterates backwards from the current bar to identify the most valid swing high/low to commence from, but it still is subjective at best. That's where the good ol' brain makes things very obvious...but try getting an EA to do it.

I have had to give up using the SDC for automation purposes but have developed a few proxy measures that seek to achieve the same thing but are simpler for EA coding. The more I am investigating this diversified technique, the more I am coming to the conclusion that the technique itself is just a means to an end and that the secret sauce lies in diversification, whether it be diversification by product range across asset classes, system diversification such as a combination of trend following and mean reverting systems, or diversification through timeframe selection.

There is not much on FF to get the teeth into regarding portfolio diversification but there are a few good books on it that I am reading at the moment but that is where my attention is being placed at the moment.
  • #444
  • Sep 14, 2015 2:21am Sep 14, 2015 2:21am
Quoting Copernicus
There is not much on FF to get the teeth into regarding portfolio diversification but there are a few good books on it that I am reading at the moment but that is where my attention is being placed at the moment.
Totally in agreement with you regarding diversification! In many forms... despite my interest in forex, I'm actually a very risk-averse person, and when it comes to funds/savings/investments, diversification is invaluable.

I'm a little curious how you're going about it, ie which platforms and risky asset classes (options/futures/cfds/etc) have worked for you, or anyone with a similar style
  • #445
  • Edited 6:00am Sep 14, 2015 5:29am | Edited 6:00am
Quoting AshR
{quote} I'm a little curious how you're going about it, ie which platforms and risky asset classes (options/futures/cfds/etc) have worked for you, or anyone with a similar style
There is lot's of publicly available info relating to performance of the Managed Futures segment that utilise portfolio backtesting solutions such as WealthLab or preferably RightEdge (if you are a serious programmer) and most of the larger players in this segment place their trades using regulated banking platforms offered by institutions such as Saxo Bank or JP Morgans etc.

However the product offerrings available today is progressively becoming more diverse to the extent that wide diversification across asset classes such as hard commodities, soft commodities, treasuries, forex, equity indices etc. can now be achieved with CFD's or EFT's (albeit at a slightly higher transaction cost). The principle aim of diversification is to flatten the equity curve and allow for enhanced returns with lower drawdowns. Diversification across un-correlated assets provides this opportunity and this is why futures offerrings have historically been such an attractive vehicle for those seeking to balance their portfolio....but now with the emergence of new product offerrings in the CFD space, retail traders like ourselves can now obtain exposure to cross asset opportunities utilising fairly simple platforms such as MT4 without having to acquire specialised platforms or technologies.

For example two brokers at the moment that are definately catching my eye in this regard (here in Australia) are IG Markets and avatrade. As discussed earlier in this thread their product offerrings are extensive and provide significant cross asset diversification exposure via some innovative extensions of their CFD/EFT range of products. IG Markets only recently now offers continuous contracts in soft and hard commodities utilising CFD's which is a boon for those traders seeking access to these assets without having to worry about futures expiry dates and rollovers.

....anyway, it is a work in progress at the moment and I also would be very interested to hear from others on FF how they go about portfolio management.
  • #446
  • Sep 14, 2015 6:39pm Sep 14, 2015 6:39pm
Quoting AshR
which platforms and risky asset classes (options/futures/cfds/etc) have worked for you, or anyone with a similar style
A bit of bad wording on my part, sorry - I actually meant in terms of performance of the platform! Although I'll have to look into the portfolio backtesting solutions (Once I'm done with going through the toptradersunplugged transcripts - thank you for the link.)

The emergence of CFDs interests me, since they're such an affordable way to get into commodities, but because they're such new offerings, there doesn't seem to be much in terms of feedback from users. In addition, the method(s) of calculating the underlying asset worries me a little. Maybe someone will chime in with their experience...

And how funny is it, that overnight, once again, we've got a new PM
  • #447
  • Sep 14, 2015 7:12pm Sep 14, 2015 7:12pm
Quoting AshR
{quote} A bit of bad wording on my part, sorry - I actually meant in terms of performance of the platform! Although I'll have to look into the portfolio backtesting solutions (Once I'm done with going through the toptradersunplugged transcripts - thank you for the link.)
Sorry mate. I should have read closer. At the moment I am using a bandaid solution for my system which comprises MT4 for the platform (due to automation requirements) and an Access database to collate portfolio data and run scenario tests with various portfolio configurations. It is a home grown solution but is sufficient to analyse and reverse engineer what the heavyweights appear to do. Given you are getting through 'Top Traders Unplugged', have a listen to the MD from Chesapeake Capital. They don't apply much science to diversification and simply work on the principle of trading as great a possible share of the available liquid investment universe as possible across as many asset sectors as possible. The results I am deriving from my quick and nasty testing tends to confirm this principle, but like others it is impossible to actually emulate what Chesapeake does due to finite capital constraints. Others like Campbell and Co. apply a lot more science to diversification and are selective in their instrument choices using historic volatility to guide them which gives hopes for me and I am only just starting to get my head into this.


Quoting AshR
{quote} The emergence of CFDs interests me, since they're such an affordable way to get into commodities, but because they're such new offerings, there doesn't seem to be much in terms of feedback from users. In addition, the method(s) of calculating the underlying asset worries me a little. Maybe someone will chime in with their experience...
Also the history for these instruments is minimal making it difficult to backtest the efficacy of trend trading using them.....but when you cross-compare price action between the futures and their derivative CFD instrument it appears to be almost identical, so I think they are likely to offer some significant benefits to those retail traders desiring broad product diversification......time will tell I suppose but it is best to be initially sceptical.


Quoting AshR
{quote} And how funny is it, that overnight, once again, we've got a new PM
Who would have thunk it!!!!......It is a bit like musical chairs at the moment.....but I must admit, I am extremely happy with the outcome. I am hoping Turnbull's vision translates into action and improved business confidence. We are lucky living in this glorious land and it is such a waste if we don't capitalise on it.
  • #448
  • Edited 12:56am Sep 18, 2015 12:31am | Edited 12:56am
Monthly Summary - August 2015 Results

Below is the monthly update for various approaches described in this thread comprising the DTT, EDTT and EDTTI (no counter-trend).

Finally...some solid results for August attributed to the very volatile sessions that saw new trends emerge in the portfolio. As with all medium to long term trend following approaches, volatility is our friend. Long may it last. We need to have some very solid months ahead to achieve similar performance results to the prior two calendar years. Keep fingers crossed that volatility persists otherwise this is just going to have to be a calendar year we chalk up as pretty b*%$#dy average.

Enjoy....with a chill session to keep you company. :-)

C

Inserted Video


Attached Image (click to enlarge)
Click to Enlarge Name: D1 Trends A.PNG Size: 310 KB


Attached Image (click to enlarge)
Click to Enlarge Name: D1 Trends B.PNG Size: 277 KB


Attached Image (click to enlarge)
Click to Enlarge Name: D1 Trends C.PNG Size: 311 KB


Attached Image (click to enlarge)
Click to Enlarge Name: D1 Trends D.PNG Size: 312 KB


Attached Image (click to enlarge)
Click to Enlarge Name: August 2015 Performance.PNG Size: 108 KB


Attached Image (click to enlarge)
Click to Enlarge Name: August 2015 Performance Chart 1.PNG Size: 54 KB


Attached Image (click to enlarge)
Click to Enlarge Name: August 2015 Performance Chart 2.PNG Size: 35 KB


Attached Image (click to enlarge)
Click to Enlarge Name: August 2015 Performance Chart 3.PNG Size: 35 KB
  • #449
  • Sep 21, 2015 9:00am Sep 21, 2015 9:00am
Benchmark Guidance of 'Good' Mechanical Systems

For those system developers out there who are wanting to know when to consider laying down the development tools and seriously consider commencing trading, then the book 'Beyond Technical Analysis: How to Develop and Implement a Winning Trading System' by Tushar Chande, a respected CTA, provides an indicative benchmark of system requirements that would meet the 'sufficient test' in regards to robustness of a trend following system for futures markets.

The benchmark requirements caught my eye and thought it might be of interest to those following this thread. While relating to futures trading, the principles still hold for alternative asset classes like stocks and forex. Obviously there needs to be a time in system development when you ultimately have to say, enough is enough and the following guidelines at least provide a degree of confidence in what would be regarded by the professional fund managers as sufficient.

See how your system fares against this guidance?

Attached Image
  • #450
  • Sep 21, 2015 9:05am Sep 21, 2015 9:05am
Quoting Copernicus
Benchmark Guidance of 'Good' Mechanical Systems For those system developers out there who are wanting to know when to consider laying down the development tools and seriously consider commencing trading, then the book 'Beyond Technical Analysis: How to Develop and Implement a Winning Trading System' by Tushar Chande, a respected CTA, provides an indicative benchmark of system requirements that would meet the 'sufficient test' in regards to robustness of a trend following system for futures markets.
Hey Copernicus,

What does it mean by 'Total Entry/Exit Conditions/Parameters'?
  • #451
  • Edited 10:08am Sep 21, 2015 9:37am | Edited 10:08am
Quoting Atokys
{quote} Hey Copernicus, What does it mean by 'Total Entry/Exit Conditions/Parameters'?
Hi A

He is referring to the number of variables (rules) used for an entry/exit condition such as variables to define a breakout entry/ moving average entry/ filters applied etc. Basically in 'EA' terms the key variables used to control entry/ exit.....if that makes sense. What the <=10 values suggest is that the simpler the better...in general. Simpler rules allow for a more robust strategy that can manage a range of market conditions, are anticipated to work better in an uncertain future and avoid curve fitting.

His views are that efficient systems use a bare mimum of variables/rules as an increase in parameters that must be met restricts trade opportunities (that for trend traders may turn out to be ultimate winners). A bit like Goldilocks.....not to little and not too much but....just right. In general more conditions reduce risk volatility but correspondingly impede profit performance and there obviously is a delicate balance that needs to be addressed.
  • #452
  • Edited 9:01am Oct 20, 2015 6:04am | Edited 9:01am
Hi C,

Chanced upon your thread and got completely interested in your approach, as I think you have a great thing going here. All the more appreciative of how forthcoming you are in sharing your thoughts and methodology. You mentioned about being nerdybut I'm actually liking the thoroughness of your posts; they have given me a picture of your thoughts with great clarity. I have also enjoyed the humour you interject in your posts. They may be lengthy, but it makes for a fun a good read.

I have just about finished on the DTT section, so while there are still whole lot of posts to get to and I look forward to the EDTT, I thought of giving this appreciative feedback to you first. Still new to the SDC, and just starting to appreciate and understand the power behind it.

I do have one question though, and I thought I'd check with you on this before I get into the EDTT section:

On Post # 150, the first chart that has the green dotted H1 SDC drawn from the Swing Low, and another solid yellow SDC that I would assume to be the D1 SDC. My initial thought is that wouldn't this actually deviate from your original DTT approach (as detailed in Post # 1), as the H1 SDC does not commence from the D1 SDC starting point? In which case we should not have this scenario if we had started the H1 SDC in line with the D1 SDC? E.g. The H1 SDC would more closely resemble the slope of the D1 SDC.

Also on a similar note on Post # 191, when you mentioned H1 SDC should mimic the width and characteristics of the D1 SDC. Every time I draw the H1 SDC from the commencement point of the D1 SDC, I notice that I always seem to get very similar widths between the H1 and D1 SDC, the only difference being a slight difference on the slope of the two. So I'm thinking, wouldnt the H1 SDC, given that it is drawn from the commencement point of the D1 SDC, generally exhibit the same width as the D1 SDC anyway?

Not sure if I'm missing something here. And apologise in advance in you have covered this in later posts or if it has been superseded in later posts. Great thread once again, and I hope you continue to share your thoughts. I have found a lot of substance here. Keep it up.

By the way, this thought has also crept up since I am aware of the EDTT and EDTTI, are you still using the DTT, or should we completely look at the EDTT/EDTTI now? I note the excellent track record, and would have thought it still worthwhile to continue if it still maintains its robustness.

BTW, I also don't want to distort the discussion on EDTT and EDTTI, which I assume is what these latter posts are on, so if it would be better for me to send this to you as a PM, let me know. Thanks.
  • #453
  • Oct 23, 2015 8:55pm Oct 23, 2015 8:55pm
Hi C,

Managed to read through the EDTT and EDTT1 portion plus the recap, and its been really insightful to say the least. Everything including the questions posted earlier is clearer now. Thanks and can really see the immense work youve done. Will continue to tune in to this thread.
  • #454
  • Oct 24, 2015 4:49am Oct 24, 2015 4:49am
Quoting greenfields
Hi C, Managed to read through the EDTT and EDTT1 portion plus the recap, and its been really insightful to say the least. Everything including the questions posted earlier is clearer now. Thanks and can really see the immense work youve done. Will continue to tune in to this thread.
Hi G

Great to see that you are enjoying the thread. I am sufficiently confident with this discretionary approach to say that it certainly is a robust method and as a discretionary approach achieves better overall performance results in terms of expectancy and stability of returns than the automated approaches I am working on at the moment.

I have ceased testing the approach as I am fairly time poor at the moment and am now back on MT4 as a platform so unfortunately cannot provide updated results for the various approaches. Needless to say I am very confident that the EDTT1 approach is the preferred option only trading in the direction of the primary trend.

If you have any queries or thoughts please continue to share them in this thread despite it's 'use by date' and using my experience with this approach I will try to assist in any way I can.

If I was to revert to a discretionary approach then the EDTT1 would be my preferred choice so it is nice to have a 'fall back' strategy if the fully automated versions don't deliver.

From my testing of other systematic trend following approaches I can say one thing with certainty. If you apply this approach with as diverse a portfolio as you can get across asset classes where you apply equal weighting w.r.t volatility to those asset classes, this approach will clean up when the trends occur.

If things don't work with the automated trading I will be back on board with this method and looking for a team of like minded traders experienced with this technique to collectively manage a portfolio with 24/5 coverage. Unfortunately this beast needs discretion on entry conditions and SDC placement and cannot be fully automated.

Regards

C
  • #455
  • Edited 10:41pm Oct 29, 2015 10:07pm | Edited 10:41pm
Some valuable resources relating to Systematic Trading and in particular Diversified Trend Trading

To those who may stumble upon this thread here is a great book resource.

Current personal fav's of mine are as follows:
1. Quantitative Investing - Strategies to exploit stock market anomalies for all investors
2. Building Winning Algorithmic Trading Systems
3. Trading Systems - A New Approach to System Development and Portfolio Optimisation
4. Unholy Grails - A New Road to Wealth
5. Stocks on the Move: Beating the Market with Hedge Fund Momentum Strategies
6. Following the Trend: Diversified Managed Futures Trading7.
7. Trend Following with Managed Futures: The Search for Crisis Alpha
8. Dual Momentum Investing: An Innovative Strategy for Higher Returns with Lower Risk
9. Beyond Technical Analysis: How to Develop and Implement a Winning Trading System
10. Inside the Black Box: A Simple Guide to Quantitative and High Frequency Trading
11. The Art and Science of Technical Analysis: Market Structure, Price Action and Trading Strategies
12. Evidence-Based Technical Analysis: Applying the Scientific Method and Statistical Inference to Trading Signals
13. Trading Systems and Methods + Website (5th edition)
14. The New Trading for a Living: Psychology, Discipline, Trading Tools and Systems, Risk Control, Trade Management



Also some great podcasts to get the head around on your long drives or the weekends.

  1. Better System Trader
  2. Trend following with michael covel
  3. Top Traders Unplugged ​

Enjoy

C

  • #456
  • Nov 2, 2015 2:46am Nov 2, 2015 2:46am
An interesting mathematical proof for trend followers who wonder whether it is best to set profit limits or let the trend run it's course before exit.

The mathematics explores different strategies applied to a 'drifted random walk model'. This form of model best describes the philosophy underlying the trend following approach described in this thread.

The optimal conclusion reached using this model is as follows:

For maximum expectancy: predict the direction of the trend, buy (or short-sell) the stock in the beginning according to the predicted trend direction (drift), and exit the position when the trend ends.

Regards

C
  • #457
  • Nov 2, 2015 3:27am Nov 2, 2015 3:27am
Hi, Copernicus
I'm not much into this long term diversified trend trading,(Iam scalper and action junkie) but I'm thinking that you can achieve much better results with individual stocks. You can catch some amazing trends there, especially during bear markets. I have seen stocks go up from a few bucks to over 100 and also from 100 down to a few dollars, and even cents. And if you trade them trough CFD you can use leverage on top of this.
FX is mean reverting market, so it's harder for the trend follower.
Maybe I'm wrong on this....
  • #458
  • Nov 2, 2015 3:36am Nov 2, 2015 3:36am
Quoting alphaomega
Hi, Copernicus I'm not much into this long term diversified trend trading,(Iam scalper and action junkie) but I'm thinking that you can achieve much better results with individual stocks. You can catch some amazing trends there, especially during bear markets. I have seen stocks go up from a few bucks to over 100 and also from 100 down to a few dollars, and even cents. And if you trade them trough CFD you can use leverage on top of this. FX is mean reverting market, so it's harder for the trend follower. Maybe I'm wrong on this....
Hi AO

I cut my teeth in the equities market when I commenced in this game and am likely to return to this asset class in the near future. I have had the pleasure of reading Clenow's latest book "Stocks on the Move" on some great diversified momentum based strategies that can be applied to this market which got my opportunity radar working overtime but recognise some of the shortcomings of equities, namely their strong correlations particularly in 'crisis periods'. However there obviously are a few powerful methods to harvest the nuanced opportunities in this sector. Thanks for the heads-up.

Anyway, am looking forward to investigating this sector in a bit more detail and identying the potential to engage this strategy with them. The strategy in this thread trades a number of indices, but there are some great benefits to the equity curve by including some single stocks as well so will keep an open mind.
  • #459
  • Nov 2, 2015 8:16am Nov 2, 2015 8:16am
Quote
The strategy is currently configured to actively monitor 24 products
Hey Copernicus,

Do you mind sharing the list of products you trade? Im always keen for new ideas on diversifying a portfolio.

*EDIT*

Oh nevermind, I think you posted it in post number 8.
  • #460
  • Nov 2, 2015 8:29am Nov 2, 2015 8:29am
Quoting Atokys
{quote} Hey Copernicus, Do you mind sharing the list of products you trade? Im always keen for new ideas on diversifying a portfolio. *EDIT* Oh nevermind, I think you posted it in post number 8.
Not much science in the selection Atokys, but sufficiently diversified to produce a fairly stable equity curve....so far *crosses fingers and toes* keeping in mind that diversification also lies in the 6 distinct timeframes traded with varying correlations.

I have moved on a bit with my thoughts on portfolio selection (assisted through ideas from some of the CTA's) in terms of instrument selection, timeframes traded and also the possibility for system diversification and that currently is a work in progress.
0 traders viewing now