Variable and figures concatenation

• 01/28/2020 at 7:37 PM #118175
  Roger

  Hi,

   

  I would like to try optimisation tests based on the time.

  I can define a trading hour to enter long or short but I would like to insert 2 variables in my trading times in order to optimize :

  trading hour

  defparam cumulateorders = false TradingHour = Time = a'0000' TradingHourOut = Time = b'0000' condenterlong=close<close[1] and TradingHour condexitlong=TradingHourOut IF NOT LongOnMarket AND condenterlong THEN BUY 1 CONTRACTS AT MARKET ENDIF If LongOnMarket AND condexitlong THEN SELL AT MARKET ENDIF

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

  defparam cumulateorders = false   TradingHour = Time = a'0000' TradingHourOut = Time = b'0000'   condenterlong=close<close[1] and TradingHour condexitlong=TradingHourOut     IF NOT LongOnMarket AND condenterlong THEN BUY 1 CONTRACTS AT MARKET ENDIF     If LongOnMarket AND condexitlong THEN SELL AT MARKET ENDIF

  As you can see my variables are a and b

  Thanks

   

  

  Roger

  Participant

  Topics: 9

  Replies: 93

  Been thanked: 30 times
  01/28/2020 at 8:41 PM #118178
  nonetheless

  What I would do is:

  TradingHour = a

  TradingHourOut = b

  and then set the range in the optimisation box, for example:

  a      070000 to 100000 in steps of 10000

  b      150000 to 210000 in steps of 10000

  Unless I’ve misunderstood what you want to achieve?

  

  nonetheless

  Participant

  Topics: 83

  Replies: 1235

  Been thanked: 978 times
  01/28/2020 at 8:54 PM #118180
  Roger

  That is exactly what I want to achieve.

  I have no computer tonight but I will try it tomorrow.

  Thanks for your help!

  

  Roger

  Participant

  Topics: 9

  Replies: 93

  Been thanked: 30 times
  01/29/2020 at 7:22 PM #118221
  Roger

  Your solution seems to work properly, thanks again, I am going to do some “curve fitting” 🙂

  

  Roger

  Participant

  Topics: 9

  Replies: 93

  Been thanked: 30 times
  01/29/2020 at 8:56 PM #118223
  Vonasi

  Roger wrote:

  I am going to do some “curve fitting” 🙂

  If only I’d known to buy on a Tuesday at 1432 and sell on Wednesday at 1559 then I’d be a rich man.

  At least you are going into this study knowing that there is a huge potential for data mining when using time as a variable. 🙂

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1812 times
  01/29/2020 at 10:06 PM #118226
  Roger

  I have spent more than one hour on it and even with overfitting I can barely reach a profit factor of 2 and more than 50% of winning trades… The time window 17h00-18h00 seems to be the most interesting window to work on but for now it is extremely disappointing…

  

  Roger

  Participant

  Topics: 9

  Replies: 93

  Been thanked: 30 times
• Author
  Posts