Help needed to get better indication of rising MA

• 07/17/2018 at 2:53 PM #76137
  Allonso

  <strong>1. indicator1 = ExponentialAverage[8](close) 2. indicator2 = WeightedAverage[50](close) 3. c1 = (indicator1[10] CROSSES OVER indicator2[10]) 4. 5. indicator3 = Weighted Average[50](close) 6. indicator4 = Average[200](close) 7. c2 = (indicator3 >= indicator4) 8. 9. indicator5 = Average[200](close) 10. indicator6 = Average[200](close) 11. c3 = (indicator5 >= indicator6[20]) 12. 13. 14. SCREENER[c1 AND c2 AND c3] (</strong>(close/DClose(1)-1)*100 AS "%Chg yest.")

  1 2 3 4 5 6 7 8 9 10 11 12 13 14

  <strong>1.  indicator1 = ExponentialAverage[8](close) 2.  indicator2 = WeightedAverage[50](close) 3.  c1 = (indicator1[10] CROSSES OVER indicator2[10]) 4. 5.  indicator3 = Weighted Average[50](close) 6.  indicator4 = Average[200](close) 7.  c2 = (indicator3 >= indicator4) 8. 9.   indicator5 = Average[200](close) 10. indicator6 = Average[200](close) 11. c3 = (indicator5 >= indicator6[20]) 12. 13. 14. SCREENER[c1 AND c2 AND c3] (</strong>(close/DClose(1)-1)*100 AS "%Chg yest.")

  Hi,

  I am not a coder but I try. In the above code (line-11) I have tried to find when the MA is higher than the period twenty days before. This works but there are many occasions when the MA’s are only marginally higher. This is really not what I want. Ideally I want cases when the MA is moving at a reasonably steep angle thereby providing a tradable move.

  Can anyone kindly suggest a way of achieving

  Regards  Allonso

  

  Allonso

  Participant

  Topics: 3

  Replies: 6

  Been thanked: 0 times
  07/17/2018 at 3:27 PM #76139
  robertogozzi

  Be aware that ProScreener is limited to a 254-period lookback, now you reached 220 in line 11 (200MA + 20 lookback).

  Slope angles cannot be calculated, you can find several topics dealing with this. It’s not a fault of PRT, they simply cannot be calculated.

  You might use PIPS to be sure the distance between the two is not so small. Line 11 could be:

  c3 = (indicator5 >= (indicator6[20] + 10 * pipsize))

  1	c3 = (indicator5 >= (indicator6[20] + 10 * pipsize))

  to make sure their distance is at least 10 pips.

   

  

  robertogozzi

  Moderator

  Topics: 69

  Replies: 16450

  Been thanked: 3634 times
  07/17/2018 at 3:51 PM #76142
  Roberto S.

  Hi Roberto, what do you suggest to use for stocks? tickSize * x times?

  Cheers

  R

  

  Roberto S.

  Participant

  Topics: 10

  Replies: 17

  Been thanked: 4 times
  07/17/2018 at 4:04 PM #76144
  robertogozzi

  Sorry, but I do not trade stocks.

  

  robertogozzi

  Moderator

  Topics: 69

  Replies: 16450

  Been thanked: 3634 times
  07/17/2018 at 4:11 PM #76145
  Allonso

  Thanks Robertogozzi,

  Its a shame slopes are not included, my problem with your suggestion is how I find out what te tick size is of UK stocks. I have ggogled but no luck so far. Maybe I should see if a smaller MA say 150 period will still achieve what I want.

  Regards Allonso

   

  

  Allonso

  Participant

  Topics: 3

  Replies: 6

  Been thanked: 0 times
  07/18/2018 at 10:51 AM #76216
  Allonso

  Hi All

  Following on from robertogozzi ‘s suggestion I used the changed code to test in four stock markets the:- Nasdaq, UK, German and NYSE. Using the factor (10*pipsize ) produced Zero results. Changing the factor to (0) and the results were in order (14, 4, 11, 19), using (1) and I got (3, 3, 0, 1) and using (2) results were (2, 1, 0, 1).

  So it seems the best method is to use (0) and eyeball the results making use of the colour change from Up to Down of the 200 MA. As a thought is it possible to code for this colour change as it must be calculated somewhere in the system. Perhaps some of you top coders could give it some thought.

  Regards to all    allonso

   

  

  Allonso

  Participant

  Topics: 3

  Replies: 6

  Been thanked: 0 times
  07/18/2018 at 11:08 AM #76222
  robertogozzi

  Screeners do not support GRAPHIC instructions.

  

  robertogozzi

  Moderator

  Topics: 69

  Replies: 16450

  Been thanked: 3634 times
• Author
  Posts