Diversified Trend Trading Approach | Page 187 | 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

  • #3,721
  • May 22, 2018 12:51pm May 22, 2018 12:51pm
Quoting Copernicus
{quote} Christ MP. Not sure how I missed this gem of a post. I think in the euphoria of testing DJI30 that a myopic focus tends to make my mind miss things. Yes mate....trading the equity curve is a great idea and something that we definitely will be considering. There is lots of info about this by the CTA's who use this concept particularly for trend following and it is very succesfully deployed by Funds of Funds who rebalance their fund portfolios by only allocating to FM's who are experiencing a drawup as opposed to a drawdown. We also can use...

Thanks for your input , glad to be of help.
What's draw me to this KAMA progress is the lack of discretionary which one of my big weakness since many variables come into play.
  • #3,722
  • May 22, 2018 3:52pm May 22, 2018 3:52pm
Cut your losses short and let your winners run in one image:

USD TRY from March. Risking 0.94% per position.

Reverter paid for losses in short SDC.

Attached Image (click to enlarge)
Click to Enlarge Name: cut losses short0IBspLf.png Size: 108 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,723
  • May 22, 2018 4:42pm May 22, 2018 4:42pm
Quoting PipMeUp
{quote} Would it be possible for you to post these trades in some excel/csv format?
Morning Pip. This is output from latest version of EA....with a few more instruments still to go. Hopefully there is enough data for you to work with over a very difficult TF period.

Just note the issues we currently have in relation to position sizing and the bumpy ride still from whipsaws expressed in last few posts.

We also need to factor in SWAP and slippage...plus tests are on current spread so need a more conservative adjuster.

Cheers C :-)

PS Out of action for a couple of days but will be back.
Attached File(s)
File Type: csv InstantKarma20180518.csv   56 KB | 265 downloads
  • #3,724
  • May 23, 2018 4:23am May 23, 2018 4:23am
Human touch is needed to correct Donchian lookback period.

Attached Image (click to enlarge)
Click to Enlarge Name: P9frDp3.png Size: 43 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,725
  • May 23, 2018 6:18am May 23, 2018 6:18am
Quoting Copernicus
{quote} Morning Pip. This is output from latest version of EA....with a few more instruments still to go. Hopefully there is enough data for you to work with over a very difficult TF period. Just note the issues we currently have in relation to position sizing and the bumpy ride still from whipsaws expressed in last few posts. We also need to factor in SWAP and slippage...plus tests are on current spread so need a more conservative adjuster. Cheers C :-) PS Out of action for a couple of days but will be back. {file}
I remember some years ago we had a discussion (with FXEZ I think), in a thread I couldn't find back, about trading the equity curve like MP suggests. The conclusion was that you need your equity curve to be positively autocorrelated. Many of the curves you posted seem to be so. That's why I asked for the CSV.

Let's play a game:
I track your equity with a EMA alpha=0.1 (= EMA 19). This period is a random (educated) guess. I take its first difference (the slope of the EMA). If the slope is positive I copy your next trade. If it is negative I take the opposite. The result is in the excel. The first trade is always the same as yours.

I leave the analyze to the savvy reader (you can do better).
Attached File(s)
File Type: xlsx InstantKarma-on-steroids.xlsx   202 KB | 464 downloads
No greed. No fear. Just maths.
  • #3,726
  • May 23, 2018 6:20am May 23, 2018 6:20am
You take all these trades and you don't know which one would go parabolic as Turkish Lira.

Attached Image (click to enlarge)
Click to Enlarge Name: NJgzu0Q.png Size: 145 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,727
  • May 23, 2018 6:29am May 23, 2018 6:29am
Quoting PipMeUp
{quote} I remember some years ago we had a discussion (with FXEZ I think), in a thread I couldn't find back, about trading the equity curve like MP suggests. The conclusion was that you need your equity curve to be positively autocorrelated. Many of the curves you posted seem to be so. That's why I asked for the CSV. Let's play a game: I track your equity with a EMA alpha=0.1 (= EMA 19). This period is a random (educated) guess. I take its first difference (the slope of the EMA). If the slope is positive I copy your next trade. If it is negative...
Now that's thinking outside the box!
Thank you for this idea.
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,728
  • Edited 4:36pm May 23, 2018 6:48am | Edited 4:36pm
Quoting PipMeUp
{quote} I remember some years ago we had a discussion (with FXEZ I think), in a thread I couldn't find back, about trading the equity curve like MP suggests. The conclusion was that you need your equity curve to be positively autocorrelated. Many of the curves you posted seem to be so. That's why I asked for the CSV. Let's play a game: I track your equity with a EMA alpha=0.1 (= EMA 19). This period is a random (educated) guess. I take its first difference (the slope of the EMA). If the slope is positive I copy your next trade. If it is negative...
Excellent idea Pip.

We now have a baseline *unconstrained* model under *trending conditions* which has been tested across 24 instruments albeit over a very short term but non-trending period in general..... which we will use as a benchmark for testing new additions to the strategy. This unconstrained model has simply so far pushed trade execution out of the everyday churn under the assumption that we are attempting to catch the meat of some possible longer term trends...as opposed to the noise of efficient markets. In it's current form it has lots of room for improvement but the intent was to select a candidate as a base for further improvement.

The simple assumption of using a fairly long range Donchian break seems to have already addressed significant adverse drawdown in the equity curve so far...... so we are happy with step 1 to the logic.

Now we are going to apply a few ideas that we think may further address the following issues present in this baseline model that:

1. Tests whether we need to head towards the short term or the longer term in terms of timeframe to catch alpha. We will be splitting the baseline model into two separate return streams being short term and long term in nature so we can look at the impact of timeframe on our baseline model.

2. Address sideways congestion whipsaws by progressively stepping future entries away from the Donchian break based on their prior failure rate. The idea being that we hope to reduce adverse sequence of losses by taking action to force us quickly away from horizontal reversion points. Note that this might be not advisable in relation to Pip's idea.

3. Address major adverse volatility outliers and possibly more of the damaging whipsaw impact by some stop loss mechanism. We may find that stop losses do actually help in this scenario;

4. Address the ATR position sizing issue to get far more consistency in position sizing using other measures.

Each of these processes will be tested against the baseline to see if these ideas are useful

The *unconstrained* model in its current state was more to test the efficacy of trending conditions and we are very happy with the result....but now it is down to squeezing alpha using risk management mechanisms which was a major bonus of EDTT and we would like to test with KARMA as well.

Once we have achieved a result that we think has real legs....then the next step is your idea. We will certainly test it out as we get nearer to a solution we are happy with. In a nutshell simply reverse the trade direction using the slope of the EMA....*christ*.....you are a living legend)

Attached Image


I will keep you posted on this thread regarding the progress of our assumptions used......but this is going to be a labor of love and a lengthy exercise.

Really appreciate that 'food for thought' Pip. I agree with Ramadas....this is 'out of the box' logic......very clever....and yes....dependent on autocorrelation of the equity curve evident by the smooth nature of drawdowns and drawups. :-)

PS luv the name. ...Instant Karma on steroids LOL A typical understatement by a mathematician:-)

Cheers C
  • #3,729
  • May 23, 2018 9:52am May 23, 2018 9:52am
Dear C

Just to be on the same page.

The 24 instrument results you presented a few posts back, are the result of KAMA cross (10,5,30 and 10,3,30) in two TFs
(M30 and M5), where the trades are executed on M5 according to the direction of M30.

Is it so?

T
Enter Signature
  • #3,730
  • Edited 6:59pm May 23, 2018 10:02am | Edited 6:59pm
Quoting thomasK
Dear C Just to be on the same page. The 24 instrument results you presented a few posts back, are the result of KAMA cross (10,5,30 and 10,3,30) in two TFs (M30 and M5), where the trades are executed on M5 according to the direction of M30. Is it so? T
Hehehehehe...... Pip is a clever man isn't he. I am literally picking my jaw off the floor :-)

Yes mate..exactly. Just be aware however that the use of ATR for position sizing assuming $500 risk makes lot sizes wildly change....and it will give cause for ulcers......we will work on the strat as discussed in my prior post which is more robust and less volatile and as a last measure then apply Pip's game changer. I will keep you posted.

C

PS A teaser....thanks to Pip

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 369 KB
  • #3,731
  • May 23, 2018 8:10pm May 23, 2018 8:10pm
An amazing story on the power of hallucinogens. Not to condone it in any way or form....but some of us may have experienced similar seminal moments in our wicked lives.

Attached Image
  • #3,732
  • May 24, 2018 3:34am May 24, 2018 3:34am
Portfolio Result - Instant Karma on Steroids - M5 only - 11 Instruments Tested 1 Aug 2017 - 30 April 2018

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 42 KB
  • #3,733
  • May 24, 2018 1:21pm May 24, 2018 1:21pm
Quoting Copernicus
{quote} What a cracker of a post to say goodbye with my friend. Given your penchant for leaving no footprints I am going to preserve these fantastic points by quote. I hope you don't mind. Extract from Veefx I truly feel price or any movement of it is the biggest trickster to our eyes. Right after position is entered, I consider Price to be non-existent in any measure. I just let my structured set of exists via Profit/Loss/Margin/Exposure/Drawdown mgmt functions be my guiding principle to manage my portfolio of positions per strategy (what I lack...
Thanks C for breaking the cardinal rule here :-(
I got cut off but discovered I still have access via an old Ipad so I am back for a bit. I was in Vegas last week for the BBMA and EDC so had a good break.

I forgot to include the most important diversification item in my list.

Fixed %Risk and Variable Position Sizing Diversification: I think my approach is quite unique to my overall strategy and integral part of how I keep my risk % fixed and still introduce extreme aggressive approach to explosive growth, enable quicker drawdown recovery and put my 'limited' margin to the max overdrive. It is a far superior risk/MM model imo and feel this could you help you guys in your respective journey. I use a contact 1.5% risk across all positions with a twist as I do not base any position sizing on account balance or account equity. That's conventional wisdom. I base it on a unique framework (idena came from stt's post on drawdown) which keeps track of what I call "Account Threshold" which is a value I maintain in real time based on account equity and balance curve differential, Peak Equity, Trough equity, Equity at weekly and monthly open. This allows me to get

  1. super aggressive (accelerate via stepping on gas!) when equity explodes in my favor breaking the barrier of account threshold
  2. super risk averse (apply engine brakes!) during drawdown via drawdown management function
  3. Run on idle when there is just nothing going on in the currency markets by running my "windingdown" function to reduce exposure (as part of my exposure management function) and generate "income" via scalpng

So ask my how my % risk is fixed at 1.5% and why my position size decreases every time I add new positions to my portfolio on the same pair or any other pair where the signal is generated and yet keeping my next equity risk at a level from which I can rapidly recover during drawdown from high watermark? This risk management model allows me to run up used margin all the way up to 95% and still have no fear of triggering a liquidation via margin call.

The answer lies in WHY my two 'diversified' portfolio #5 and #6 in the screenshots below (which I added in May) failed miserably. My individual strategies (#1 thru #6) relies on equity growth as step 1 to kick in all these various diversification techniques at full throttle. The weakness is the first 10% of account growth/drawdown above/below initial deposit. That is why it is crucial for me to run in idle state and generate income for the first 10% of initial account deposit. After that, it is all smooth sailing and unattended portfolio management.

No disrespect to anyone here attempting diversification techniques for a long time. I feel (although too early for me to comment as an expert) managing equity curve across strategies via diversified approach (the path you guys are taking in this thread if I understand this correctly) )is an exercise that will only lead to failure in growth. My 2 cents.... Diversification techniques that results in capping growth just to improve volatility adjusted returns is falling short on the growth curve. It might be good for income-producing strategies but I would question whether retail spot FX is a good vehicle purely for low-risk income strategies even in the professional fund management space. Options might be a better vehicle for such strategies.

Here's my MTD performance stats for May just to give an idea of what I shared above. I am also sharing how easy it is to 'manipulate' trading stats and how it can mess up the trader's psyche for those who are not advanced in their journey. I still think Returns over Max DD and Time-Adjusted Gain is the only two relevant measures of performance. Rest are all either deceiving or just obstacles in our journey. Hope this helps shed some light further on what I do.

I did not realize you can develop EAs so need to watch what I share. Probably my last post here until I return with solid long term performance to declare success in my own journey :-)

Adios Bro. Keep up the good work.

Attached Image(s) (click to enlarge)
Click to Enlarge Name: Image1.png Size: 31 KB
Click to Enlarge Name: Image2.png Size: 20 KB
Click to Enlarge Name: Image3.png Size: 10 KB
Click to Enlarge Name: Image4.png Size: 16 KB
Staying in my lane...
  • #3,734
  • May 24, 2018 1:49pm May 24, 2018 1:49pm
Quoting VEEFX
{quote} Thanks C for breaking the cardinal rule here :-( I got cut off but discovered I still have access via an old Ipad so I am back for a bit. I was in Vegas last week for the BBMA and EDC so had a good break. I forgot to include the most important diversification item in my list. Fixed %Risk and Variable Position Sizing Diversification: I think my approach is quite unique to my overall strategy and integral part of how I keep my risk % fixed and still introduce extreme aggressive approach to explosive growth, enable quicker drawdown recovery...

The ghost who never dies...

For the love of god man, stay back and keep posting from time to time...

PS: Trying to make us jealous with your vacation in vegas. All jokes aside, glad you had a wonderful break...
  • #3,735
  • Edited 11:25pm May 24, 2018 9:52pm | Edited 11:25pm
Quoting VEEFX
I was in Vegas last week for the BBMA and EDC so had a good break.
Nice V. Viva Las Vegas :-)

Quoting VEEFX
I forgot to include the most important diversification item in my list. Fixed %Risk and Variable Position Sizing Diversification: I think my approach is quite unique to my overall strategy and integral part of how I keep my risk % fixed and still introduce extreme aggressive approach to explosive growth, enable quicker drawdown recovery and put my 'limited' margin to the max overdrive. It is a far superior risk/MM model imo and feel this could you help you guys in your respective journey. I use a contact 1.5% risk across all positions with a twist...
Stunning mate. Great idea What you are talking about here is a recognition of the distinct phases of advantageous and disadvantageous market conditions (auto-correlation) that exist in a market and how to utilise position sizing to dial up or dial down your risk exposure. I don't worry about those equity curves you highlighted as they have important info in them that we can utilise.....as you do as well for the bigger picture.


Quoting VEEFX
The answer lies in WHY my two 'diversified' portfolio #5 and #6 in the screenshots below (which I added in May) failed miserably. My individual strategies (#1 thru #6) relies on equity growth as step 1 to kick in all these various diversification techniques at full throttle. The weakness is the first 10% of account growth/drawdown above/below initial deposit. That is why it is crucial for me to run in idle state and generate income for the first 10% of initial account deposit. After that, it is all smooth sailing and unattended portfolio management....
Not so sure mate....but I appreciate where you are coming from and I like your solution you propose.

Quoting VEEFX
My 2 cents.... Diversification techniques that results in capping growth just to improve volatility adjusted returns is falling short on the growth curve. It might be good for income-producing strategies but I would question whether retail spot FX is a good vehicle purely for low-risk income strategies even in the professional fund management space.
In a way we are approaching this from two perspectives. You use your position sizing risk modifier to respond to prevailing market conditions.....while I use risk weighted returns to flatten the equity curve.......and here comes the rub.....so I can increase my position sizing through multipliers that would otherwise leave the single strat guys for dead with account blow ups. There is a symmetry in what we are both trying to achieve here V in the most efficient use of our finite capital to get best bang for buck. You say Potato and I say POTATO....sort of thing going on here. :-)

Quoting VEEFX
Options might be a better vehicle for such strategies. Here's my MTD performance stats for May just to give an idea of what I shared above. I am also sharing how easy it is to 'manipulate' trading stats and how it can mess up the trader's psyche for those who are not advanced in their journey. I still think Returns over Max DD and Time-Adjusted Gain is the only two relevant measures of performance. Rest are all either deceiving or just obstacles in our journey. Hope this helps shed some light further on what I do. I did not realize you can develop...
Cheers mate and continue to flout your personal rules and contribute here on this thread mate. This thread is about sharing great ideas and pushing the limits.....and I am getting some great ideas from you mate....and hope you also get some ideas from this thread yourself. You are on a great journey mate....and kudos to you.

Cheers

C

PS V....as a personal favour please leave your posts on my thread and don't delete them. Context and following chains of thought are so important as we try to pull apart these markets and a disjointed record for me upsets the apple-cart as I continue to peruse my journey in order to bring fresh thought to the process. As I have said many times this is a mutual journey for participants in this thread and everyone is welcome to it. If we build creative thought in this thread from a range of different interpretations....then some bright spark may just take something from this process to make a wild dream of 'gedanken thought' come true.
  • #3,736
  • Edited 2:19am May 25, 2018 2:07am | Edited 2:19am
Quoting Copernicus
{quote} Nice V. Viva Las Vegas :-) {quote} Stunning mate. Great idea What you are talking about here is a recognition of the distinct phases of advantageous and disadvantageous market conditions (auto-correlation) that exist in a market and how to utilise position sizing to dial up or dial down your risk exposure. I don't worry about those equity curves you highlighted as they have important info in them that we can utilise.....as you do as well for the bigger picture. {quote} Not so sure mate....but I appreciate where you are coming...
Thanks man. Appreciate the kind words. In all honesty, this is all part of my master 'exit' plan to participate in this thread. It was decided when I left the Edge in single candle thread earlier this year. Your thread focuses on Risk mgmt and it just felt appropriate to conclude my thoughts here as a departing gift for those who can think and take it forward. Anyways, I want to comment in this underlined paraphrase as my post has huge significance to reduced risk so want to make sure there is crystal clarity on what I came here to convey. In all fairness, I was trying not to share as this might truly be my edge within the grand scheme of things and everything I have shared thus far is played out in a well choreographed synphony within my Universal Time Action strategy.

"In a way we are approaching this from two perspectives. You use your position sizing risk modifier to respond to prevailing market conditions.....while I use risk weighted returns to flatten the equity curve.......and here comes the rub.....so I can increase my position sizing through multipliers that would otherwise leave the single strat guys for dead with account blow ups. There is a symmetry in what we are both trying to achieve here V in the most efficient use of our finite capital to get best bang for buck. You say Potato and I say POTATO....sort of thing going on here. :-)"

My principle behind Position Sizing as part of RM/MM comes with a point of view of keeping it a constant. By this I mean, market conditions, open positions, equity or balance growth, drawdown or any other such condition should have NO impact on position sizing. It is viewed as an independent function within the context of applying the "universal rule of law". May I remind you Markets are Universal and Neutral (deep meaning in these words!). Under this principle, any measurement or calculation we do on variables by definition becomes a variable. By this I mean, one cannot achieve consistency in the long run if the underlying foundation of core principles is based on ever-changing rules of engagement. By this I mean, one has to think long and hard about how one part of strategy component could have a equal and opposite reaction to another part of strategy component (Newton's 3rd law!). The concept behind my Universal Entry is the epicenter of "everything remains fixed i.e. constant in a ever-changing homogeneous, hybrid and organically vibrating market environment. So with this context in mind, let's revert back to what I said about fixed %Risk based position sizing. But first a little more context (as I am in mood to share my long hours of research on currency markets)....

(Please take no disrespect as I am writing this from the viewpoint that noobs can also understand and not just advanced participants of this thread)

No one would be trading FX if there was no leverage allowed. So every account is a margin account. Every dollar deposited comes with a purchasing power to buy fiat currency (please note I did not mention buy and sell... there is only BUY involved... ask me how because platforms like MT4 does a fantastic job of hiding the real inner-workings of currency trading making it appear so easy for noobs to get onboard by clicking a few buttons)
Now, Account deposit = Trading capital = Total Available Margin
Now comes the spicy sauce.... Every Pair has a "Required Margin" component as part of their trading conditions specific to the broker. Broker can change these trading conditions at will and with very little or no notice. Just an email saying "we are changing margin requirement for xyz currency pair due to increased risk in market conditions" screwing most retail traders... big scam to choke portfolio traders and long term position traders who love to carry their positions forward are now in extreme dilemma! So please ask yourself how can we prevent this type of market condition risk with conventional risk based position sizing or kelly formula or fixed lot sizing or whatever Van Thorp recommends in his book (I have not read ANY books from Gurus!). You simply cannot! Why? Because either you are basing your position sizing on account balance or account equity and BOTH are ever-changing components of your strategy. House of Cards will always crumble when wind blows from any direction.

So I ask again. How can we protect ourselves from such a massive manipulative uncontrolled unregulated broker-dependent environment and have a fair fighting chance to succeed with consistency in the long run?

EDIT: Also ask yourself, what do ATR/ADR and Required Margin for any given pair have in Common? Ask yourself, how I can manage to allocate a fixed amount of risk per position universally across my portfolio of positions with absolutely ZERO regard to mathematically enforced correlations or daily ranges or volatile pairs or non-volatile pairs. These all mean nothing to me nor do I care!

Allow me to mention two Math geniuses on FF... TwoBlink and Master-Mind. The answers lie within these threads
Staying in my lane...
  • #3,737
  • May 25, 2018 2:43am May 25, 2018 2:43am
From Hot to Cold - The 2nd Law applied to testing strategies

Attached Image


Damn....The Euphoria of the 'On Steroids' autocorrelated solution dissipates with the testing phase of applying the principle. Place a rain check on ordering those Bugatto Chiron's just yet guys....Houston has found a problem.

You may recall that Pip got hold of the data I supplied and intrigued us all with his finds on auto-correlation using a novel "reverse trade approach' dependent on win/loss sequence.

The returns were stellar and salivating....and in a similar manner I did the same thing using a similar applied approach to get this.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture1.JPG Size: 86 KB


Could it be possible. Had we found the 'grail'?

Well the devil as always is in the details.

Have a look at the trade sequence to understand what I mean.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture2.JPG Size: 146 KB


Now the principle we applied was the position of reversing the sign of the trade based on the preceding result.....and this is exactly what we did....but here comes the kicker...........

We should have noted the entry date and time as well as we were trading a multi-strat timeframe approach on M5 and M30 whereby on a Donchian break we released two trades at once. It was therefore in practice impossible to use this theory against the practice as both trades taken had to be of the same directional sign.

Now the impact of all this was that we were biasing the results to produce these outstanding returns.

Here is the real story on M5 using the same approach....but this time correctly for a single timeframe which brings us down to the ground with a crash.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture3.JPG Size: 85 KB


Now while we may be crying into our cold tea at the moment....it still raises an interesting point and one definately worth considering.

Based on principles of autocorrelation....this 'steroid' approach works....provided the sequence of profit data is correct. What we are after is either multiple consecutive losses or multiple consecutive profits for the principle to shine. What we don't want however is up...down....up....down....up ....down sequences that have the reverse effect on the steroid solution.

Just to demonstrate that it works....here is an example of this approach applied correctly to another strat.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture4.JPG Size: 74 KB


Once you get this concept in your head it is hard to shake as it provides a relatively new angle to pursue for a diversified portfolio.

Now refer back to the prior chart to look deeper at the equity curves of the systems and the timing of their ups and downs.

I have included a blend result which is a composite of adding the steroid to the raw karma result. Now look at what happens.

1. The M5 Karma Raw result is a divergent strategy on M5 which had a positive CAGR of 22% over the period (1 Aug 2017 to 30 April 2018). Now this is symptomatic of the fact that on balance there were better divergent conditions on this timeframe over this period. For example...January was volatile and this trend following strategy thrived. Herew I was saying that this period was a tough one...but conveniently forgot January.
2. The M5 Steroid result was a loser over this period. Why is this considering that the prior raw Karma performed favorably over this period? Well the reason is that autocorrelation was sporadic and restricted to January. This has a big impact on the Raw Karma due to it's open ended profit nature which produced stellar profits on a restricted period that paid for the balance of losses for the balance of the period.....however the trade sequence wasn't so kind to Steroid...where the windfall was not as large in January to pay for the balance of non-correlated periods.
3. Now look at the blend and do not be dismayed by the short time frame result.....but understand how the principle is working so we can step into the next crucial point.

If autocorrelation exceeds the noise over the longer timeframe.....then the blend is going to detect is as a bias. The reason for this is that the outliers picked up by the raw strategy and going to be cancelled by the outliers of the 'on steroids version' and positive autocorrelation is going to amplify the returns. Noise however is going to persistently decay the raw result and the steroid result.....so in applying this principle you need to consider where the greatest chance that noise won't apply.....and I can give you a hint....when markets are in crisis.

Soooooo it is stil very worthwhile pursuing this approach and considering the development of numerous strats as over the long term....the blend is going to smooth the returns of the overall portfolio.

Cheers

C
  • #3,738
  • Edited 3:06am May 25, 2018 2:50am | Edited 3:06am
Quoting VEEFX
{quote} Thanks man. Appreciate the kind words.
Hehehehe. We were both having Tolstoy moments at the same time. Cheers for this mate. Fantastic stuff which I am going to absorb for a while.

PS Just to demonstrate this universal law in action in regards to position sizing for this portfolio approach. I test every market using a fixed dollar amount of $500 per trade which is simply a desired calc of a hypothetical starting % of trade capital of $200K.

The fixed dollar amount scales every instrument the same way in terms of fixed dollar value. Once testing is complete I have an array of equity curves which I blend to get a $ volatility signature across the entire portfolio. I then determine what is my total risk appetite in terms of drawdown and then I apply the same multiple to all instruments in the portfolio to universally scale to that desired tolerance......so I see where you are coming from in terms of 'universality'.

In the same way I never cherry pick instruments and provided they meet liquidity requirements.....any liquid instrument needs to be put through the same hoop as it is the universal principles that are important for sustainable outcomes in an uncertain market context.

Cheers V....will continue to absorb cause there is so much worthwhile info in it.



C
  • #3,739
  • May 25, 2018 7:27am May 25, 2018 7:27am
Back on Track looking at Divergence

We have taken a recent journey into the land of autocorrelation within equity curves and have got some great info from it that we will store away for the future.....but now we are reverting back away from our inspired deviation into the familiar world of divergence.

I have already demonstrated that on M5 we have a successful strat for the period concerned...so we will continue to fine tune and progress.

So it is not a waste of time meandering as we design and the process is likely to give us multiple potential strategies to apply for the system blending from the KAMA indicator itself with a few tweaks.

Cheers

C
  • #3,740
  • May 25, 2018 8:58am May 25, 2018 8:58am
Inserted Video


Inserted Video


LOL
Inserted Video


Inserted Video


Inserted Video
2 traders viewing now