Pathfinder swing TS
Forums › ProRealTime English forum › ProOrder support › Pathfinder swing TS
- This topic has 2,004 replies, 6 voices, and was last updated 1 year ago by Gianluca.
-
-
01/14/2017 at 5:53 PM #2110401/14/2017 at 6:34 PM #21112
Hi Guys,
For time reasons I focused on all backtests for the next period Jan2 (16-31) and have verified your work and moved the best backtests (Rating A) to the hall of fame. Thanks again for your support and I hope it gave you a little fun. I guess some of you have dealt with instruments that they never have traded so far.
The next step is to discuss how to manage the bunch of backtests. I have seen that Pfeiler already work on this task for Robin’s Russell 2000 backtests. I would prefer to have only one or two robots for each instrument. Suggestions are welcome.
We had only one signal for the first period and the silver robot delivered (at least in life mode) a profitable trade. I will create a spreadsheet and list all the signals/trades because and that is the good thing of this idea the signals repeat every year with a high probability. We have 13 promising setups for the next period. We will see.
Good trades and Best,
Reiner
01/15/2017 at 7:46 AM #2114901/15/2017 at 6:44 PM #2119001/15/2017 at 8:38 PM #2120601/15/2017 at 9:14 PM #2121101/16/2017 at 1:53 PM #2131101/17/2017 at 7:51 AM #2140001/17/2017 at 9:10 PM #21509Based on Robins work on the US Russel (and of course Rainers PF), I have created a PF Daily program for Silver fot the whole year, that incorporates the seasonal pattern configuration for each period.
It delivers a better result than the “Pathfinder Daily V2 Silver” program I had before, but there is still room for improvement (e.g. low % Winning Trades).But this should be just a initial start, to have a single program for every instrument.
I hope Rainer has time to check, if this is sth we can work with and improve the numbers for every season which is coming.I have also tracked all details in an online excel sheet, and will continue to do so, hopefully with results, you guys will work out:
https://docs.google.com/spreadsheets/d/1pBoKX44kX_GSzHNc0CgueKdZAuWBCnauoK_bh8ht6j8/edit?usp=sharing(FY = Full Year)
5 users thanked author for this post.
01/17/2017 at 10:58 PM #21517Hey Pfeiler,
really nice work. thank you!
how did you optimize the program? each seasonality one by one? and then put it in the complete system? or did you optimize it in the complete system?
and in your excel sheet, the max drawdown in may is 15,9%, but in the complete years it is smaller. any idea why this is the case?
I am looking forward to try this for other instruments.
01/18/2017 at 8:49 AM #21521@flowsen: Great, that you want to give it a try.
I started, by removing the seasonal patterns for Feb-Dez, so that only Jan is left. Then I optimised the numbers for Jan (J1+J2), transfered em to the spreadsheet, and continued to Feb by increasing the number from 1 to 2 (If CURRENTMONTH = 2), then “rinse and repeat”. It really didn’t work out to optimize for every period, I optimized for every month. I also skipped the months, where Rainer already came to the conclusion, that there is no gain (see spreadsheet on the first page). But this is a new code, maybe worth to check again. If you have hammered out all the numbers, reincoporate them in the code with seasonal patterns for Jan-Dez.
To avoid multiple work .. which instrument do you want to have a look at?
@MichiM: Didn’t want to steal your Silver optimization away. I just read that you had it optimized for the whole year, but then couldnt find the PRT files. So I started from scratch, but maybe you can check the numbers from my program with yours. I hope, we came to the same conclusions. 😉
01/18/2017 at 9:03 AM #2152301/18/2017 at 9:18 AM #2152401/18/2017 at 9:47 AM #2153401/18/2017 at 10:40 AM #21538@ Pfeiler. I will start with Soybeans and Dax, since I allreay did some optimization for that and they look really promising.
1 user thanked author for this post.
-
AuthorPosts
Find exclusive trading pro-tools on