Optimization moving average crossing strategy

• 03/21/2020 at 12:00 AM #122782
  nonetheless

  Jan wrote:

  To avoid comparing the same MA

  That’s great Jan, should speed things up too, 70 less pairs to check.

  I think that only has to be added to one set of the MAs, no? ie not the MAtypeV2 as well

  I have tried altering the period but so far 15 on both sides seems to give the best result.

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/21/2020 at 12:16 AM #122783
  Francesco

  Best optimization for 200k:

  ONCE MAType = 3
  ONCE MATypeV2 = 9

  1 user thanked author for this post.

  nonetheless

  

  Francesco

  Participant

  Topics: 32

  Replies: 439

  Been thanked: 200 times
  03/21/2020 at 12:23 AM #122786
  nonetheless

  Excellent – thanks Francesco!

  That combination does well enough in the ancient history, but more importantly it’s getting good results this year — 3 and 9 it is.

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/21/2020 at 12:27 AM #122787
  nonetheless

  @Jan — forget my query about the alteration to the code, of course it only has to be added once at the beginning of the list. Brain fog.

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/21/2020 at 10:06 AM #122790
  Jan

  Hi Nonetheless,  thanks for the reply, not sure what you meant ” @jan — forget my query about the alteration to the code, of course it only has to be added once at the beginning of the list. Brain fog.”

  Have a nice weekend.

   

  To avoid comparing the same MA (assuming the periods remain fixed), which is then meaningless, the very simple code is:

  //Before all average definitions:

  if MAType <> MATypeV2 then

   

  //after all average definitions

  Endif

  You have to do this code AROUND all averages, both MAType and MATypeV2 in my opinion.

  I think that if you code it only around the MAType defined averages, the first set,  the strategy optimizing would give an error because when both (MAType and MATypev2) are the same,  MAType would be excluded and undefined, which results in Nothing (MAType) crossing a average MATypeV2, which should give a error.

  

  Jan

  Participant

  Topics: 10

  Replies: 90

  Been thanked: 29 times
  03/21/2020 at 10:54 AM #122794
  nonetheless

  Jan wrote:

  not sure what you meant

  Because my first thought was that “if MAType <> MATypeV2 then” had to be added to each MA — ie 140 times, but of course it’s just once at the beginning of the list. That was the foggy brain part. Never mind… have a great weekend!

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/21/2020 at 1:26 PM #122821
  mr blue

  DAX 1H looks very promising. Thanks

  

  mr blue

  Participant

  Topics: 10

  Replies: 77

  Been thanked: 14 times
  03/21/2020 at 3:31 PM #122832
  nonetheless

  Here’s another way of using this. The attached itf isolates just the ‘search engine’ part so that it can be inserted into any existing strategy. For example, if you had certain indicators and conditions on a daily TF, you could insert this on a 1h TF and it will find a pair of crossing MAs to use as additional entry conditions within the context of your other filters and parameters etc.

  Very effective!

  2 users thanked author for this post.

  Monochrome, GraHal

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
  03/21/2020 at 7:41 PM #122867
  frenqle

  Hi Jan,

  This one I am trying to activate on the DAX 1 min. It shuts down after a few seconds, minutes or hours (without reason).

  I hope you can find what I’m missing. Alvast bedankt!

   

  

  frenqle

  Participant

  Topics: 3

  Replies: 8

  Been thanked: 0 times
  03/21/2020 at 8:17 PM #122869
  GraHal

  Has it ever traded … even once?

  What is the Rejection stop message when it shuts down … you can see these in the Orders List, Rejection / Cancelled Tab.

  

  GraHal

  Participant

  Topics: 203

  Replies: 8373

  Been thanked: 2024 times
  03/21/2020 at 9:11 PM #122872
  frenqle

  It never traded.. I quits without reason..

  

  frenqle

  Participant

  Topics: 3

  Replies: 8

  Been thanked: 0 times
  03/21/2020 at 9:54 PM #122873
  GraHal

  Try below … as without the Flatbefore (which you didn’t have) then all times are flat after … if you understand my logic?

   

  DEFPARAM cumulateOrders = False // Cumulating positions deactivate Defparam flatafter = 160000 //close the position at 16:00 (day trading) /could be switched off Defparam flatbefore = 090000 //you don't have this in your code SET TARGET pPROFIT 150 once StartE = 90000 //start time for opening positions once StartL = 110000 //ending time for opening positions (only trading in the morning)

  1 2 3 4 5 6

  DEFPARAM cumulateOrders = False // Cumulating positions deactivate Defparam flatafter = 160000 //close the position at 16:00 (day trading) /could be switched off Defparam flatbefore = 090000 //you don't have this in your code SET TARGET pPROFIT 150 once StartE = 90000  //start time for opening positions once StartL = 110000  //ending time for opening positions (only trading in the  morning)

   

   

  

  GraHal

  Participant

  Topics: 203

  Replies: 8373

  Been thanked: 2024 times
  03/21/2020 at 11:00 PM #122876
  Jan

  hi Frenqle,

  hard to test in the weekend.

  What do mean with ” after a few seconds, minutes or hours (without reason).” you tried different timeframes ? If you had used the hourly timeframe, did it then stop immediately ?

  It seems consistent with the base code.

  Did you try to leave the original code and set it fixed to the required 2 averages, does that work ?  (I saw you stripped to code to only the 2 averages you want)

  KR Jan

  

  Jan

  Participant

  Topics: 10

  Replies: 90

  Been thanked: 29 times
  03/22/2020 at 12:10 AM #122879
  dnystrom

  Would there be possible to make a code that look at a higher TimeFrame pair crossing, and when that happened execute an order on a lower time frame for an optimized crossing for that timeframe? Or would that just result in very few trades? Just been doing ProRealTime for a week so thats out of my competence level =)

  What I want to do is, find a optimal crossing for a long position on lets say the daily or 4H chart, and when that happened, look for a optimized crossing for long position on a much lower timeframe like 15 min, 5 min, etc. So we just take trades in the “right” trend at the moment. And wise versa for the short position. That should improve the winning %?

  Awesome code both Jan and Nonetheless, thank you!

  

  dnystrom

  Participant

  Topics: 1

  Replies: 18

  Been thanked: 6 times
  03/22/2020 at 12:31 AM #122884
  nonetheless

  @dnystrom I dont think you would want an actual crossover in both time frames as that would be a fairly rare occurrence and, as you say, result in v few trades. Better would be to use some kind of trend indicator in the daily TF, then insert the part of the code i posted above (MACross finder) in the shorter TF and let it find the best crossover to work in conjunction with your daily filter.

  It’s possible of course to load it in both TFs and optimize them separately, but I wouldn’t expect it to work very well. (or maybe it would – try it and let us know!)

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1237

  Been thanked: 978 times
• Author
  Posts