Diversified Trend Trading Approach | Page 182 | 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

  • #3,621
  • May 17, 2018 4:33am May 17, 2018 4:33am
Quoting Copernicus
Houston we have a problem From initial heads exploding....we are now in this position :-) {image} He who won't be named has developed an EA based on the logic we have applied to the Instant Karma and it just doesn't stack up over a range of market conditions. A false start.....not unusual as we were blinded by our short term samples we tested manually which turns out to be cherry picked and suitable to current market conditions :-( {image} Sorry guys...can't release the EA but I have attached the fidelity checker between our manual tests and the...
1. Use 1h
2. After MA cross put pending order over last candle high/low or last (x) candles low/high
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,622
  • Edited 5:16am May 17, 2018 4:42am | Edited 5:16am
Quoting thomasK
Dear C One of the biggest problems we face trying to implement KAMA in Forex markets, is the same nature of those markets. Kaufman , when designed KAMA, had the stock market in mind. See below a weekly diagram of NVDA using just our two KAMA combination. Obviously it works perfectly with a profit of more than 1500% on the last signal only. But the stock market is a trending market. Forex is a mean reverting market. KAMA is a great tool if applied to the correct market. {image}
Cheers T. It is not done and dusted yet. I am shaking things up a bit and getting some interesting results. I will share them as I go and then we can get an EA to put it through the wringer. All good mate. Strategy creation takes a lot of time and evaluation. I don't want to find that I threw away a potential goldmine just by applying better sound logic. I have a littany of wasted backtests from the old days when I would just test till I was blue in the face until an equity curve appeased me. These days I apply logic in the design to avoid having to look for a needle in the haystack.......and if I could resurrect those old wasted backtests now....I would find that many had legs but I just could not see the woods for the trees :-)

The amended version I am on now is multi-timeframe using M30 for trend direction and M5 for entries and exits. It is producing some great results and may be worthwhile testing through EA. We are definitely after a return stream that has no resemblance to the current strategies we deploy and a periodic sniper in scalp mode may just be the ticket.....but it needs to be divergent with lots of positive skew and low kurtosis for risk management purposes......Unlike what you may hear elsewhere....'catching the falling knife and letting risk build with unrealised losses' doesn't work for me as a long term solution. Diversification is the way to avoid this dillema facing retail traders who inevitably deploy convergent strategies to make their bread and butter...for a short time only.

My main emphasis at this point in time is as follows: The KAMA cross has no problem trading the trend.....the issue is with the whipsaws and their material nature.....so my logic is directed towards this issue at the moment. Down the track I can look at long only or short only options and combos and apply them to equities if necessary if that is the solution. We are asset neutral mate and will consider any addition to the portfolio that delivers a potential edge. If artworks or sports betting had liquid derivatives we would be considering them as well....*sort of*....but you get the picture.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 188 KB
  • #3,623
  • May 17, 2018 4:44am May 17, 2018 4:44am
Quoting Copernicus
Portfolio Results to 30 April 2018 - Fixed Universe Both EDTT and Reverter went marginally backwards on very little trade volume. In contrast R I think you might be having a little snigger with your dynamic universe for the month. We need a system to plug Aprils weakness R. We have a long mean reverter and and long/short breakout....now we need a plug in the short reversion space. Have a close look at market conditions in April and see what possibly works on the fixed universe selection. Take a little chance...it don't mean nothing. Sometimes you...
Dynamic portfolio with 2 positions maximum in sub-sector yielded such results:

Attached Image
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,624
  • May 17, 2018 4:48am May 17, 2018 4:48am
Quoting Ramadas
{quote} 1. Use 1h 2. After MA cross put pending order over last candle high/low or last (x) candles low/high
Mate....can you please have a crack at this while I crunch out another version I am looking at. Let me know the results on EURUSD for 1 Jan 2018 to 28 Feb 2018 as I am simply trying at this moment to identify potential material variations in the equity curve that may help us.

C
  • #3,625
  • May 17, 2018 4:49am May 17, 2018 4:49am
Quoting Ramadas
{quote} Dynamic portfolio with 2 positions maximum in sub-sector yielded such results: {image}
I thought you had nailed it mate. Great result :-)
  • #3,626
  • May 17, 2018 4:53am May 17, 2018 4:53am
Quoting Copernicus
{quote} Mate....can you please have a crack at this while I crunch out another version I am looking at. Let me know the results on EURUSD for 1 Jan 2018 to 28 Feb 2018 as I am simply trying at this moment to identify potential material variations in the equity curve that may help us. C
Well, EA doesn't work on 1h.... I am trying to hire programmer which will make EA working in every TF
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,627
  • May 17, 2018 5:00am May 17, 2018 5:00am
Quoting Ramadas
{quote} Well, EA doesn't work on 1h.... I am trying to hire programmer which will make EA working in every TF
Mate....that is being addressed as we speak. He just punched out a quickie on M30 to stop me bugging him. The plan is for a large suite of testing variables for multiframes, filters etc dedicated to KAMA. Won't be long. Meanwhile we work on the manual backtester to inform him of various approaches to consider. Feed your ideas through me and then I can project manage the process with him without too much overload from us. To keep us all in sync please focus on ideas and test the clappers out of the EA on receipt according to the purpose it was designed. Eg. M30 for the last one. Many more variations coming. It will keep us all occupied as a strategy factory and not twiddling our thumbs.:-)
  • #3,628
  • May 17, 2018 5:17am May 17, 2018 5:17am
Quoting Copernicus
{quote} Mate....that is being addressed as we speak. He just punched out a quickie on M30 to stop me bugging him. The plan is for a large suite of testing variables for multiframes, filters etc dedicated to KAMA. Won't be long. Meanwhile we work on the manual backtester to inform him of various approaches to consider. Feed your ideas through me and then I can project manage the process with him without too much overload from us. To keep us all in sync please focus on ideas and test the clappers out of the EA on receipt according to the purpose it...
OK, thank you for organizing this. I will backtest manually some other markets.
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,629
  • May 17, 2018 5:27am May 17, 2018 5:27am
LOL

Attached Image (click to enlarge)
Click to Enlarge Name: LJ664nX.png Size: 1.4 MB


https://www.marketwatch.com/story/th..._share_twitter

Intelligence in society:

Attached Image
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,630
  • May 17, 2018 5:40am May 17, 2018 5:40am
Quoting Copernicus
{quote} Mate....can you please have a crack at this while I crunch out another version I am looking at. Let me know the results on EURUSD for 1 Jan 2018 to 28 Feb 2018 as I am simply trying at this moment to identify potential material variations in the equity curve that may help us. C
OK, I will try this range and market.
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,631
  • May 17, 2018 5:41am May 17, 2018 5:41am
Quoting Ramadas
LOL {image} https://www.marketwatch.com/story/th..._share_twitter Intelligence in society: {image}
Bwahahahahahahaha :-)

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 34 KB


They should launch it on FF. Imagine the number of hits :-)
  • #3,632
  • May 17, 2018 5:50am May 17, 2018 5:50am
Attached Image (click to enlarge)
Click to Enlarge Name: kxnoQyk.png Size: 469 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,633
  • May 17, 2018 6:03am May 17, 2018 6:03am
Since we throw KAMA ideas on the table, my previous post concerning NVDA gave me the following idea.
As we can see in NVDA, during those last 4,5 years price never even touched KAMA 10,5,30
So apart from looking for KAMA crossovers, how about using just price crossovers with KAMA 10,5,30
ON TWO TIMEFRAMES (otherwise we should go back to elementary school ).
Entry according to the direction of the higher TF, entries and exits on lower TF.
Here is a relevant screenshot of DOW.


tk

PS The area in the rectangle gave no signal since we place a sell stop order on H4 which was never triggered
Attached Image (click to enlarge)
Click to Enlarge Name: 2018-05-17_1358.png Size: 357 KB
Enter Signature
  • #3,634
  • May 17, 2018 6:14am May 17, 2018 6:14am
Quoting thomasK
Since we throw KAMA ideas on the table, my previous post concerning NVDA gave me the following idea. As we can see, during those last 4,5 years price never even touched KAMA 10,5,30 So apart from looking for KAMA crossovers, how about using just price crossovers with KAMA 10,5,30 ON TWO TIMEFRAMES (otherwise we should go back to elementary school ). Entry according to the direction of the higher TF, entries and exits on lower TF. Here is a relevant screenshot of DOW. tk PS The area in the rectangle...
Awesome mate. Good idea. It is in the project pipeline. I will post results on the forum and details when we get to it. Just working on finalising version 1 and putting version 2 to the test. Yours will be up next.

T...Do you have my email addy?
  • #3,635
  • Edited 6:32am May 17, 2018 6:16am | Edited 6:32am
Quoting Copernicus
{quote} Mate....can you please have a crack at this while I crunch out another version I am looking at. Let me know the results on EURUSD for 1 Jan 2018 to 28 Feb 2018 as I am simply trying at this moment to identify potential material variations in the equity curve that may help us. C
Was it supposed to be a difficult period? Because it wasn't....

1. Position 1 atr daily
2. 1h tf
3. entry last high/low after signal
4. risk 1% compounding

Attached Image (click to enlarge)
Click to Enlarge Name: 1OXYNbs.png Size: 45 KB


Attached Image (click to enlarge)
Click to Enlarge Name: QCkTK6V.png Size: 49 KB


Attached Image (click to enlarge)
Click to Enlarge Name: Kby0l1f.png Size: 174 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,636
  • Edited 6:43am May 17, 2018 6:26am | Edited 6:43am
Quoting Copernicus
{quote} Bwahahahahahahaha :-) {image} They should launch it on FF. Imagine the number of hits :-)
Shhhh, don't give them ideas..... and FBI will close FF!

Attached Image
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
  • #3,637
  • May 17, 2018 7:01am May 17, 2018 7:01am
Quoting Ramadas
{quote} Shhhh, don't give them ideas..... and FBI will close FF! {image}
LOL :-)
  • #3,638
  • May 17, 2018 7:04am May 17, 2018 7:04am
Quoting Ramadas
{quote} Was it supposed to be a difficult period? Because it wasn't.... 1. Position 1 atr daily 2. 1h tf 3. entry last high/low after signal 4. risk 1% compounding {image} {image} {image}
It is the long range tests that nails it with whipsaws. This particular version is very susceptible to changing market conditions.....anyway thanks for this mate. We will do a few long range tests on the EA when we get it with your settings to check whether it stands up. :-)
  • #3,639
  • May 17, 2018 7:26am May 17, 2018 7:26am
Quoting Copernicus
{quote} Thanks for this Pip. I am wondering whether the application of an SDC of a finite bar interval might be a better tool than ATR for volatility adjusting position sizes. If I apply say a 10 bar SDC (for this thought experiment only) with a defined 1.0 Standard deviation width at each discreet time interval (bar) and then measure the distance in pts between upper and lower bars and then standardise by tick (pts) value for universal use, this might more comprehensively attack volatility. I could then develop a rolling indicator for each progressive...
In a trending market the price gets far from a MA. If the lag of your MA is constant at all frequencies (linear phase) and if the trend is constant (straight line) you can show that the distance is proportional to the trend. Even if it is not the case, you can estimate the mean distance of the price to the MA and its variance (in a rolling window). Because the MA lags below/above the price by the average value you can adjust by adding the mean to the MA. Surround the MA with 2 and 3 SD. The slope of the (original) MA over SD (slope/sd) gives an estimate of the signal noise ratio to decide (subjectively) if the price is trending. Color code up=green/down=red/flat=black. You get the pic below. Perhaps useful to trail a SL or to enter on breakout.
FYI: the MA is adpatative to the frequency spectrum. It's not a KAMA.
Attached Image (click to enlarge)
Click to Enlarge Name: GBPUSD-H4.png Size: 45 KB


BTW did you get my email?
No greed. No fear. Just maths.
  • #3,640
  • May 17, 2018 7:31am May 17, 2018 7:31am
Instant Karma - Version 2

It will be interesting to see how this holds up in the long term and across instuments. This is an intraday scalper of momentum with no stops and open profit condition which uses M30 for trend direction and enters and exits on M5 using crossovers.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 204 KB


Attached Image (click to enlarge)
Click to Enlarge Name: Capture2.JPG Size: 130 KB


Attached Image (click to enlarge)
Click to Enlarge Name: Capture3.JPG Size: 101 KB


I'm done and dusted for the day.

Last drinks. Cheers C

Inserted Video
5 traders viewing now