ATR miscalculating!

• 06/10/2018 at 10:13 AM #72752
  Despair

  I think I might have come to an understanding (maybe). When I look at the IG chart, I think IG shows for a future just the raw data of the front months (there is also very limited history) while PRT somehow also for the futures shows a continuous contract. PRT obviously splices the contract together from the front months and the differences in price we see are a consequence of the chosen rollover method.

  But does this really explain the differences? Shouldn’t be at least the last days (the front month) be the same prices while older prices are affected by the adjustments at rollover?

  

  Despair

  Blocked

  Topics: 97

  Replies: 1034

  Been thanked: 218 times
  06/11/2018 at 12:12 PM #72871
  Despair

  I’m still fighting with this problem and the future rollover can’t be the only reason since the problem is also present when I look at CFD contracts that don’t expire and therefore don’t have any rollovers. The deviations in price for CFD seem to be smaller than for futures but they are there.

  Also I still can’t see how rolling over should affect the most recent prices that should come from the front month. That older data might be back adjusted I can understand but last weeks data for example should be the same, shouldn’t it?

  For the deviations of the CFDs I’m completely clueless. From both IG and PRT so far no answer in this matter. 🙁

  
  @Nicolas: Do you have any idea?

  

  Despair

  Blocked

  Topics: 97

  Replies: 1034

  Been thanked: 218 times
  06/11/2018 at 3:48 PM #72910
  Nicolas

  Data displayed are the same if you set your custom trading hours like in the attached picture.

  On their website, IG displays the value data in London time. So if you want the same display, you must configure the market on New York time and use the option “apply settings above to non-intraday timeframe”.

  1 user thanked author for this post.

  Despair

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/11/2018 at 7:58 PM #72934
  Despair

  @Nicolas: Thank you very much for looking into this! But I have a hard time to understand. IG uses London time, this makes sense, IG is british. But why do I have to choose New York time in PRT to make it similar. I need to understand this because otherwise I can’t incorporate it in my programming. The prices I see on IG’s platform I also get from IG’s API.

  Then I also don’t understand how the timezone affects for example chicago wheat’s close. Wheat has no extended session and closes every day for several hours, so the closing prices for the last days should be the same no matter what timezone, or am I wrong?

  

  Despair

  Blocked

  Topics: 97

  Replies: 1034

  Been thanked: 218 times
  06/11/2018 at 8:32 PM #72937
  Nicolas

  I dont know. I mark this topic as resolved.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/16/2018 at 1:40 PM #73431
  Vonasi

  Despair wrote:

  Sure, I just never thought about trying other smoothings when it comes to the ATR.

  Stumbled across this today:

  https://www.prorealcode.com/prorealtime-indicators/average-filter-regression/

  You could try 66 different styles of averages and see which gives the best result!

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1811 times
• Author
  Posts