Screener close < low[1] does not work

• 08/23/2017 at 12:06 PM #44428
  xlim

  I am trying to find securities where the close of the current bar is lower than the low of the previous bar. Unfortunately the simple code below does not produce correct results. I have screened the US Tech 150 list using weekly and daily time frames. It seems like the screen is completely ignored. Hope someone can help.

  c1 = (close < low[1])

  SCREENER[c1]

   

   

  

  xlim

  Participant

  Topics: 1

  Replies: 4

  Been thanked: 0 times
  08/23/2017 at 1:52 PM #44446
  Nicolas

  Please be sure to have live datas and not “End Of Day” ones for that shares list. Otherwise the results could be delayed and that’s why you think that it doesn’t work the way it should.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  08/23/2017 at 3:46 PM #44477
  xlim

  Hello Nicolas,

  thank you for your fast reply. If I understand you correctly, the screener does not use weekly or daily data points like weekly O/H/L/C even when I set the screener to weekly data. Otherwise the screener would work. But this to me looks like a software bug.

  Because even if I subscribe to real time data it will still not access the correct data points of the weekly data. It would just use the newest data received from the real time data feed.

  I hope there is a definite way to access O/H/L/C of weekly or daily data. Please explain – may be I am missing something.

  Regards

  xlim

  

  xlim

  Participant

  Topics: 1

  Replies: 4

  Been thanked: 0 times
  08/23/2017 at 4:10 PM #44480
  Nicolas

  The OHLC are correct in EOD feeds or live ones. Your screener is working fine for me with NYSE shares list with live subscription, results are correct and in accordance with your code. What is the shares list you believe is not working?

   

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  08/23/2017 at 4:53 PM #44482
  xlim

  Hello Nicolas,

  You are correct – most of the shares I looked at are correct. I don´t know why I thought that more of them are screened incorrectly. Anyway I went through my list and found the following two shares that should not be included in the list. I used a weekly time frame. The close of week 33 is not lower than the low of week 32. Do you have any idea why these two are included in the screen?

  OSK
  UTX

  Regards

  xlim

   

   

  

  xlim

  Participant

  Topics: 1

  Replies: 4

  Been thanked: 0 times
  08/23/2017 at 9:23 PM #44498
  Nicolas

  In order to help you, please give me the name of the shares list so I could try to replicate your issue on my own.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  08/24/2017 at 7:16 AM #44510
  xlim

  Hello Nicolas,

  sorry about that. Here are 3 stocks and their corresponding lists. As you can see, the close of week 33 is not lower than the low of week 32.

  HCI = US NYSE Stocks Weekly
  SUNS = US Nasdaq Stocks Weekly
  GAIN = US Nasdaq Stocks Weekly

  Regards
  xlim

  

  xlim

  Participant

  Topics: 1

  Replies: 4

  Been thanked: 0 times
  08/24/2017 at 9:56 AM #44512
  Nicolas

  I do not have HCI in my result window for NYSE stocks. Do you have live data for NYSE shares?

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  08/24/2017 at 11:38 AM #44523
  xlim

  Hello Nicolas,

  This is very strange but the tree stocks are not shown in my lists anymore. I should heave taken a screen shot.

  Currently I am not using live data. But I just noticed that your results for the NYSE stocks are completely different from what I am getting.

  You have 243 matches and I get 818 matches. Also the displayed stocks are totally different. Please take a look at the enclosed screenshot.

  Regards
  xlim

  

  xlim

  Participant

  Topics: 1

  Replies: 4

  Been thanked: 0 times
  08/24/2017 at 2:41 PM #44551
  Nicolas

  You are using prorealtime shares data while I tested with CFD  contracts with IG feeds. That’s why we have these differences. So we are not testing the same way 🙂

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
• Author
  Posts