Diversified Trend Trading Approach

EDTT - Preliminary Results of Medium Range Tests - Discretionary Systems

.....continued from post 1952

We left our portfolio in post 1952 with 5 instruments looking at ways to plug the weakness in the drawdowns between March 2013 to October 2013 and between October 2016 to February 2017.



Given our knowledge of the relatively uncorrelated nature of equities to our existing portfolio (dominated by forex pairs and precious metals), our assumption was that maybe the addition of the Dow Jones (DJI30) to our diversified portfolio might assist in further amplifying CAGR without adding any further drawdown heat.

So let's see how we did under this assumption.

Let's first refer to the performance metrics of the prior portfolio.



We were getting excited using a multiplier of 12x as we were able to lift our CAGR to 59.0% with a max draw of 19.5% for the period 1 Jan 2010 to 16 August 2017. Surely it doesn't get better than this?.....Perhaps we are in for a nice surprise.

Let's first look at the performance of the DJI30 in isolation.

DJI30 (Dow Jones)

Annual Results



Summary



Commentary
Now it is easy to get complacent and assume that the performance results in isolation are no big deal....and hence drop that instrument and look for something with bigger bang for buck......but you need to be very aware that it is how this instrument contributes overall to portfolio robustness which is the critical ingredient.....and in this regard we need to remember that what we were searching for was an instrument that could fill the drawdown gaps. From such humble beginnings look what this individual contribution of DJI30 does to the overall performance metrics of the portfolio.

Portfolio Blend - Currently only 6 instruments

Annual Results



Summary



Through the simple addition of what appeared to be an average performer in isolation....we have accelerated our CAGR from 59.0% to 76.8% for the entire 7 year period while retaining our max drawdown to approximately 20% (19.1%) using a position multiplier of 14x.

That's it.....take some time to digest this as it is a BIG DEAL.

Just look at the Gross annual returns. Can you see any red in there? Sure we need to subtract SWAP and slippage from these results....but there is more than enough joy to spread around.



So where to now. What new instrument do we add to the mix to hopefully improve things further?

Let's review the individual returns of the portfolio to search for clues.



The DJI30 certainly nailed the weak patches of the equity curve....but there is room for more drawdown reduction between October 2013 and December 2013......mmmmmmm........more equities.......looks like we are getting in a FTSE kind of mood.


Stay tuned......

6 instruments down now only 18 to go.

Cheers

C

Now that's a nice breakout on USD SEK. Something happened to USD, but don't care what. Almost 2.5 daily volatility. Probably some a-hole said something in Jackson Hole /tic.

How about some grains or interest rates? Does your broker have them? Or carbon emissions. I can help.

Bang!!!!.....You nailed that R. Hopefully there are more a-holes in Jackson hole and USDSEK has a chance to visit hell

I am using the Dukascopy universe at the moment....but yes I am in if you can find reliable data we can share. :-)

EDTT - Preliminary Results of Medium Range Tests - Discretionary Systems

.......continued from post 1962

Recap
We left our portfolio in post 1962 with 6 instruments assuming that we could further fill the drawdown gaps by adding another instrument that was correlated to the equities market. We already had 1 equity instrument (DJI30) but we needed an additional contribution from another to help smooth the equity curve of the portfolio which was hitherto dominated by forex pairs and precious metals. Our choice for inclusion was randomly selected amongst the equities indices that were available on Dukascopy......and our choice was the good ol' FTSE. Was it a mistake or not?

This post is particularly important to get across the following ideas that are essential for portfolio blending and I will be spending some time on it:
1. Firstly do not assume that just because an instrument delivers a negative overall return that it might not assist in stabilising the volatility of returns of a broader portfolio. This post will demonstrate that as a broad assumption you actually do not need each individual return stream to have a positive slope. The key take-out to focus on is the correlation contribution the entire return distribution makes to the overall portfolio and NOT necessarily the profit contribution.
2. Our focus is on risk management and NOT profits. Remember how in a previous post we discussed that for a market that has fat tails and does not conform to a gaussian distribution, by simply strictly managing risk at all times and letting profits run....you are virtually guaranteed with the Law of Large Numbers that in the longer term you will be profitable. Follow this principle and you will not be disappointed. Remember that the vast majority of traders out there cannot get this into their skulls and that is why trend following/momentum will endure in the long term....no matter what the deniers will say.

Ok....now let's get into it.........*with a bit of music to keep you entertained while scratching your heads thinking....is this guy just a BS artist*




Let's first refer to the performance metrics of the prior portfolio.




We were flying high using a multiplier of 14x to produce a CAGR of 76.8% with a max draw of 19.1% for the period 1 Jan 2010 to 16 August 2017. Why bother diversifying any further with results like these....I hear you cry? Well despite the euphoria we have the following to deal with:
1. Our trade sample is still far to low (501 trades over 7 years or about 1 trade every 5 days on average) for my liking to give me the required confidence to commit large capital towards this strategy. I need to increase the sample size significantly to make me more convinced that this exercise has not been a cherry picked one (hindsite bias or selection biased).
2. For the purposes of deploying this strategy we are neutral in our preferred choice of instruments. Our only requirement is that it is tradable and highly liquid. The assumption that a particular market is more suited for mean reverting strategies or trending/momentum strategies is a red-rag to a bull for those with an understanding of market complexity and statistics. What may not be trending today, last year or 5 years ago.....may on a hairpin change market condition to a trending one. Remember it is the fat tails that drive us....not the normal noise of the market.....and fat tails occur unpredictably in any market irrespective of it's apparent 'normal' day to day machinations. We may just have been lucky in our choice of 6 instruments for the portfolio from a vast potential universe to select from. It is only through greater diversification and the Law of Large numbers that any statistical aberration from selection bias will be smoothed out. Once again refer to a biased coin. How many tosses does it take to determine the coin has a bias? Well there is no finite answer apart from the tendency for the bias to be detected as the sample size increase. This is the principle we need to take to our portfolio construction.

.....anyway I am digressing.....and back to the portfolio results.

So now we want to include FTSE. Let's see how it performs with this strategy in isolation first.

FTSE

Annual Results


Summary



Commentary
OMG.....yuck...you can't be serious Copernicus......what a dog.......that's a loss maker if ever I saw one?

Don't be so fast to judge Poindexter....remember we said it is how the instrument contributes to the portfolio and not it's performance in isolation. Ok....so here are a few points to digest.
1. First we unfortunately did not have data available from Dukascopy for the entire period. The best I could drum up was data from about 19 Sep 2011 onwards and we needed a history available before we could commence trading so my start was in about July 2012. No problemo....that's just how the cookie crumbles and in the real live world of trading this portfolio we would be adding new instruments to the portfolio on an ongoing basis.
2. Yes....it has performed poorly over the period and here is why. Have a look at the weekly price chart below. Fortunately I do have long range (less reliable) data from Avatrade.

You can clearly see that the period of the back-test endured very unfavourable market conditions for trend following/momentum based systems due to the noise of the price series and frequent whipsaws. This does not mean however that FTSE does not offer trend following/momentum opportunities. For example refer back on the chart below to the blue highlighted period which would have been excellent for applying trend following/momentum based systems.

Now despite the poor performance over the period, our focus on risk management has confined the potential risk. This will pay off in spades when we look at the portfolio result.

Unfortunately to make a decision to not trade the FTSE is actually a hindsight statement and needs to be avoided. For a portfolio, the decision to include or not include an instrument is based on correlation factors as opposed to performance factors a typical retail trader may be used to.

Attached Image (click to enlarge)

So how did FTSE contribute to the overall portfolio despite it's apparent abysmal performance in isolation.

Portfolio Blend - Currently only 7 instruments

Annual Results


Summary



Commentary
Now closely compare the results above with the smaller portfolio of before (with only 6 instruments). Despite the inclusion of a poor performer and applying a 14 x position size multiplier across the entire portfolio the CAGR was 74.0% (previously 76.8%)...so a slight reduction in $....but here is the kicker....the max draw was only 17.3% (previously 19.1%). So pound for pound in terms of the MAR ratio (CAGR/Drawdown) our risk weighted return is now 4.29x as opposed to the previous 4.02x.........heheheheheheheheh. I bet you didn't think that would occur?

What in effect has happened is that the *relatively* small losses associated with FTSE have less detrimentally affected the entire portfolio that the *relatively* large gains in reduced drawdown......nifty eh!

Let's now have a look at the individual returns of the portfolio as an overview.



It is starting to get visually messy but the important thing to understand from a visual perspective is how they overlay and interact with each other to produce the single equity curve of the consolidated portfolio.

Ok.....so these explanations are starting to get lengthy....but hopefully you are starting to get the idea in how to produce some stunning portfolios. My future posts in relation to growing this portfolio will simply state the overall results each time a new instrument is tested.....and I will only focus on important matters are they arise as opposed to these long winded lengthy blow by blow explanations.

Stay tuned......

7 instruments down now only 17 to go.

Cheers.....may your portfolios prosper. :-)

C

Food for Thought

Attached Image

 

1. Anxiety, adrenaline and automation
2. Trend-Following with Valeriy Zakamulin: Testing Profitability of Trading Rules (Part 6)
3. An Introduction to Pairs Trading on Quantopian
4. The Definitive Guide To Momentum Investing and Trading
5. Trends Watch: August 24, 2017
6. We will get volatility wrong at transition periods - Need to prepare before the event
7. Don’t Touch the Computer
8. How the Next Quant Fund Crisis Will Unfold
9. Trading the Trend Podcast – EP007 – Letting Winners Run
10. Top Traders Round Table – Abrams | Karahasanoglu | Lo #09
11. Top Traders Round Table – Abrams | Karahasanoglu | Lo #10

Enjoy

C

I think I will have to get my hands dirty in importing data from LCG to the simulator...

USD TRY fell even more (3 times usual volatility), I am already in profit mode after 2.5R in 2 days

Great stuff R. That is going to be a windfall

EDTT - Preliminary Results of Medium Range Tests - Discretionary Systems

.......continued from post 1967

Well I did say in post 1967 that I would keep it brief. In this post we include the addition of USDJPY to the portfolio blend. Let the results speak for themselves. :-)

........with a little music.



Annual Results


Summary


Portfolio Composition Return Profiles



Stay tuned......

8 instruments down now only 16 to go.


C

Time for some relaxing music ....

There is always time for that :-)

Am happy to have discovered this thread. I have read up to post #100 and am still reading. It seems the thread has really evolved from its early days. I want to start applying this stretedy but hope there is no much diversion from the early ideas of the system.

Thanks
Copernicus an awesome system.

Hi i :-)

Thanks for the kind words of support.

The logic surrounding the the breakout techniques (DTTR and EDTT) and variants has been very consistent apart from a few simplifications and the need to strictly monitor the tendency to over-trade. DTTR is a new kid on the block described in more recent posts that looks hopeful as a worthy additional to the system arsenal.

I have been inspired with recent back-testing to further commit to these approaches given their eye bulging performance metrics over the medium term.

What we are really trading here is momentum and with our recent back-testing....keeping things down to the lower timeframes. While we refer to these approaches as being 'trend following', the nuances that apply here that is different to traditional trend following is that we are trading momentum under shorter term swing moves of broader longer term trends.

There is no subjectivity in trading momentum. It either has directional bias of a particular desired momentum (eg. p = mv) or in terms of our trades (delta p/delta t which is defined visually by the SDC) or it doesn't. As soon as it doesn't we are out always at a lower loss than our initial stop. No ifs or buts.

Take it slow i as there is lot's to read in this thread and ensure you rigorously back-test yourself to get in the swing of things before you attempt to launch live.

Great to have you on board the gravy train :-)

C

EDTT - Preliminary Results of Medium Range Tests - Discretionary Systems

.......continued from post 1973

In this post we include the addition of USDSGD to the portfolio blend.

........with a little music.




Annual Results


Slight deterioration of risk adjusted performance metrics such as MAR.....but heh....that's the world of probabilities. We are still cooking with gas


Summary




Portfolio Composition Return Profiles



Stay tuned......

9 instruments down now only 15 to go.


C

Visual Logic of EDDT showing a few missed trades and a king hit - A Walk-through of the EDTT

I thought I would help out those just starting out getting into this thread with a visual chart update of the the most recent logic applied to the EDTT. As mentioned with i in post 1977, there have been a few amended nuances to an otherwise consistent strategy.

So what better instrument to use for an example of powerful momentum moves than XAUUSD (spot gold).

So it is early May 2011 and we have had a few wins under our belt and things are looking good on M30. Note that we are keeping things on M30 going forward. You can use other timeframes to help you interpret current price action....but to keep it simple.....keep to trading one timeframe only and make this M30 as this is what I am testing on and we have stats to back us up.



Pending Order Placement

We have just been closed out of a moderate swing move long on 4/5/2011 and I am looking to place my pending orders for a new trade. Where do I place them?



Remember we are looking for the fat tailed moves....not the normal machinations (ups and downs of market noise). I want a significant material price move under momentum before I get interested again.

As I scroll back on mychart there is nice blue sky long above the last swing high on the chart but I want to stretch things higher to avoid any nasty false breakouts so I place my pending order entry at 1604.241. There is a swing low back at 30/3/2011 and likewise I place my pending order short well below the swing low to avoid false break-downs.

Now notice how the distance to the pending order long entry and pending order short entry is almost equidistant......well 9 x out of ten....unless a prior swing low or prior swing high is VERY obvious......this is a good place to put your pending order levels. You therefore need price to really stretch either long or short from current price levels to be triggered. This is how I avoid the noise.

Time Moves forward


Nothing happening at the moment....just the incessant market noise. No clear signs of momentum happening between my pending order levels aside from a slight long drift.

Time Moves forward


zzzzzzzzzzzzzzzzzzzzz.....*burp*....zzzzzzzzzzzzzzzzzzzz.......nothing to see here apart from perhaps the commencement of a possible swing long move from a swing low point defined by your SDC with a 1.0 width.



Time Moves forward - Sweat is starting to bead on the forehead and the trigger finger is itchy - Preparing for Setup



Ok....so now it has heated up and we are about to be triggerred. What I do now is confirm my longer term swing move (A) from the prior obvious swing low .....and also look for the possibility of a shorter term swing move with higher momentum than A.....let's call this B. For me to have a B, I need to have sufficient bars of data (visually) to strongly suggest it is a shorter term swing with strong momentum on a longer term swing of lower momentum. This therefore increases my chances that momentum is rising at the point of entry (without having to look at volume measures) and thereby give me two bites at the momentum cherry.

Remember that sometimes we can be right with our momentum direction but wrong in terms of our projection of momentum going forward. I therefore (where possible) have two bites at the momentum cherry with two different momentum projections. Sometimes you cannot find a second shorter term momentum swing. In that case....don't force it and try to create one. Just let price action confirm this for you using the SDC set at 1.0 width.

So now let's draw our projections from the highest high at the bar immediately prior to entry and set our initial stops and trailing stops.



Ok....so now we have prepared our setup and waiting for entry (hopefully on the next bar)....otherwise each bar you adjust again....and again.....and again.....or if momentum breaks down and clearly reverts....then start again....no trade.

Trades Activated


Bang!!!!!! we are active. Now remember as we ride the trades that we only extend our SDC's to the prior bar's last high....each and every time a new high is made. This forces us to continuously adjust our trailing stops.

Let's step forward a bit.

Trade Active - Step forward

Attached Image (click to enlarge)

The shorter term momentum move is about to be stopped out as the trailing stop moves up the projection.......damn!!!! But we have stringently managed our loss in that we have reduced our loss $ from initial entry as we trail our stops on the projected past momentum. What this means is that momentum is drying up on the shorter term swing......but we still have a final bite of the cherry with our longer term momentum move. The fat lady is yet to sing on that.

The trailing stop on the longer term momentum move has only etched up a notch from the initial stop placement.....Ok......*phew*....*wipes brow*


Trade Exit - Short Term Momentum Move


Bwahahahahahaha....*sobs*......that's how thee cookie crumbles. Crosses fingers for long term momentum move.

Trade Active - Step forward
We are still active on longer term momentum move and it's not looking good.......about to be stopped out. We reached a new high with the SDC and the trail was adjusted in accordance with the new high......but.......Bwhahahahahahahaha.....mummy doesn't love me :-)




To be continued..........


C

A Walk-through of the EDTT......continued


Ok so we are about to be hit with a close on the longer term momentum move. Let's see if that eventuates.





Keep it together Copernicus....just rinse and repeat......back to setting up new pending order entries :-)

Pending Order Placement




Pre Setup - Defining Momentum Swing Moves


Note how we have clearly defined the longer term momentum from the prior obvious swing low (A). We try to define a shorter term faster momentum move to assist our trade entry....but in this instance no such momentum move exists. Note that however I draw the shorter term momentum swing it does not have a significantly different momentum direction than the longer term momentum swing. In this case....we don't force a new higher momentum swing and just accept a single trade outcome with the longer term momentum.

Preparing for Setup

Attached Image (click to enlarge)

Trade Active


Note how we have just escaped being knocked out again with the trailing stop.....but now we have a new high and need to reset the SDC. Hopefully we stay alive until our next action at 2.5 x R.

2.5 x R in Profit

Attached Image (click to enlarge)

R is the distance between the initial stop and the entry price. At 2.5 x R we are 3.5 x R away from the initial stop or 2.5 x R away from entry. In this instance R = 31,372 points. Therefore at 110,502 points away from the initial stop we want to reduce the width of the SDC from 1.0 to 0.5.

The reason we do this is to not leave to much profit on the table when momentum breaks down.

So adjust the SDC once 2.5xR is triggered to 0.5 SDC and reset the trailing stop on the amended SDC.




......to be continued......

C