Pathfinder Trading System

• 01/06/2017 at 6:02 PM #20082
  Mark

  Reiner,

  Hope you are well,

  Just a quick question regarding seasonality, how have you calculated this? How much data do you have? I don’t deny it improves the results but i’m a bit concerned that it makes the code curve fitted as we cant predict the future, would it not be better to fix 2 contracts across all seasons instead of 1-1-0-3-2-3-1 etc.?

  Mark

  

  Mark

  Participant

  Topics: 4

  Replies: 103

  Been thanked: 17 times
  01/08/2017 at 4:37 PM #20198
  Reiner

  Hi Guys,

  I have reworked all Pathfinder daily versions and have attached a Pathfinder daily V2 template and an excel with all required data including the seasonal adjustments. I have found some additional instruments such as Cotton, ASX, SMI and SAF. I have removed EURUSD because of the poor backtest results.

  The rating approach is the next step that I want to improve because as already discussed it’s not really comparable because of the different data histories. I think a better approach is to limit the history to 15-20 years.

  By the way I use the results of this analysis for the Pathfinder swing idea.

  Best, Reiner

  2 users thanked author for this post.

  Robin, winnie37

  

  Reiner

  Participant

  Topics: 3

  Replies: 435

  Been thanked: 1095 times
  01/08/2017 at 5:09 PM #20207
  Reiner

  Hi Mark,

  The seasonal adjustments of the Pathfinder DAX 4H version are the result of the backtest from Aug 2010 until  December 2016. On the equity clock webpage you will find a DAX statistic for the last 20 years. Here you can see that for instance  the probability in April, October, November and December is 70% and higher. Nobody knows how the prices will develop in the future but we can check the behavior in the past to find a statistical advantage – isn’t a 70% probability not a good setup  to increase the position size?

  Pathfinder monitors the price trend with the help of the signal line and will deliver a trade signal when the system detect a breakout with a statistical advantage.  The seasonality mechanism is an additional step to calculate the position size for a quality trade signal. The system is still profitable with almost all settings and it’s your personal decision if you trade with one, two or more contracts.

  I have compared the values with the result of the maximum available IG data history from 1991 compiled with an Pathfinder daily version and you can see the risky months. May be it’s a good approach to reduce the size  in the risky month to 1.

  ONCE January1 = 3 //0 risk(3) ONCE January2 = 0 //3 ok ONCE February1 = 3 //3 ok ONCE February2 = 3 //0 risk(3) ONCE March1 = 3 //0 risk(3) ONCE March2 = 2 //3 ok ONCE April1 = 3 //3 ok ONCE April2 = 3 //3 ok ONCE May1 = 1 //0 risk(1) ONCE May2 = 1 //0 risk(1) ONCE June1 = 2 //1 chance ONCE June2 = 2 //3 ok ONCE July1 = 3 //1 chance ONCE July2 = 1 //3 ok ONCE August1 = 1 //1 ok ONCE August2 = 1 //3 ok ONCE September1 = 3 //0 risk(3) ONCE September2 = 0 //0 ok ONCE October1 = 3 //0 risk(3) ONCE October2 = 2 //3 ok ONCE November1 = 2 //1 chance ONCE November2 = 3 //3 ok ONCE December1 = 3 // 1 chance ONCE December2 = 2 //3 ok

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

  ONCE January1 = 3 //0 risk(3) ONCE January2 = 0 //3 ok ONCE February1 = 3 //3 ok ONCE February2 = 3 //0 risk(3) ONCE March1 = 3 //0 risk(3) ONCE March2 = 2 //3 ok ONCE April1 = 3 //3 ok ONCE April2 = 3 //3 ok ONCE May1 = 1 //0 risk(1) ONCE May2 = 1 //0 risk(1) ONCE June1 = 2 //1 chance ONCE June2 = 2 //3 ok ONCE July1 = 3 //1 chance ONCE July2 = 1 //3 ok ONCE August1 = 1 //1 ok ONCE August2 = 1 //3 ok ONCE September1 = 3 //0 risk(3) ONCE September2 = 0 //0 ok ONCE October1 = 3 //0 risk(3) ONCE October2 = 2 //3 ok ONCE November1 = 2 //1 chance ONCE November2 = 3 //3 ok ONCE December1 = 3 // 1 chance ONCE December2 = 2 //3 ok

  Cheers, Reiner

  3 users thanked author for this post.

  Reddi, reb, Mark

  

  Reiner

  Participant

  Topics: 3

  Replies: 435

  Been thanked: 1095 times
  01/09/2017 at 5:56 AM #20246
  winnie37

  Thanks for the huge job you did!!! I will look at it very closely 🙂

   

  

  winnie37

  Participant

  Topics: 30

  Replies: 167

  Been thanked: 16 times
  01/09/2017 at 1:05 PM #20255
  Elsborgtrading

  One short position opened in pathfinder V6 DAX 4H @11532.2

  3 users thanked author for this post.

  Mark, Alco, Brage

  

  Elsborgtrading

  Participant

  Topics: 37

  Replies: 359

  Been thanked: 154 times
  01/09/2017 at 1:42 PM #20258
  dajvop

  I confirm, though at 11531.0 for me.

  

  dajvop

  Participant

  Topics: 3

  Replies: 598

  Been thanked: 501 times
  01/09/2017 at 1:49 PM #20263
  ALE

  @11532 for me

  1 user thanked author for this post.

  Mark

  

  ALE

  Moderator

  Topics: 17

  Replies: 569

  Been thanked: 223 times
  01/09/2017 at 2:14 PM #20266
  Mark

  short at 11532 for me

   

  

  Mark

  Participant

  Topics: 4

  Replies: 103

  Been thanked: 17 times
  01/09/2017 at 2:22 PM #20267
  winnie37

  same thing 🙂

  

  winnie37

  Participant

  Topics: 30

  Replies: 167

  Been thanked: 16 times
  01/09/2017 at 2:51 PM #20269
  traderfred

  Same thing, 11532..but only live, no position if I run the backtest..

  Is it the same for u?

  

  traderfred

  Participant

  Topics: 4

  Replies: 42

  Been thanked: 13 times
  01/09/2017 at 2:53 PM #20270
  Mark

  Why is that?

  

  Mark

  Participant

  Topics: 4

  Replies: 103

  Been thanked: 17 times
  01/09/2017 at 4:23 PM #20290
  Mat_CH

  Yes same for my 11532 at 1:00PM…and in backtest not happened.

  Who have reson…i don’t know if the best is close to position or not…someone have a response?

  Thanks

  

  Mat_CH

  Participant

  Topics: 1

  Replies: 17

  Been thanked: 11 times
  01/09/2017 at 4:33 PM #20291
  winnie37

  It’s surely not the best allocation, but is it a good idea to have a panel of several daily pathfinder working together, and stop the inactive when 2 are in position with smaller account?

  For example, 15 daily pro-oder pathfinder daily instrument working with a 20k account to profit from the maximum exposition .

  When 2 are on position, the others are stopped until one is free.

  What you think of that? is it a stupid money management?

   

  

  winnie37

  Participant

  Topics: 30

  Replies: 167

  Been thanked: 16 times
  01/09/2017 at 4:54 PM #20293
  ALE

  @traderferd

  You should see the position on probacktest next candles.. at 17:00 (UTC/GMT +1)

  1 user thanked author for this post.

  traderfred

  

  ALE

  Moderator

  Topics: 17

  Replies: 569

  Been thanked: 223 times
  01/09/2017 at 6:28 PM #20313
  traderfred

  @winnie37: that’s what i’m doing 🙂

  I have a small account, with 4 bots running and don’t want to have several instrument open at the same time..

  So in the afternoon, I removed the 3 others and let DAX running.

  1 user thanked author for this post.

  winnie37

  

  traderfred

  Participant

  Topics: 4

  Replies: 42

  Been thanked: 13 times
• Author
  Posts