Announcement

Collapse
No announcement yet.

Partner 728x90

Collapse

merge policy questions

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    merge policy questions

    Hi
    I trade ES. I was wondering what kind of policy I should use for accurate back testing. The default is MergeBackAdjusted which uses an offset value to make a chart continuous. So I was wondering if this would render my back testing inaccurate because the chart has been adjusted.

    Or should I use DoNotMerge?


    Thanks in advance.

    #2
    Hello sys_trader,

    Thank you for your post.

    This is really up to you on which Merge Policy to use for futures. If you want to test on the raw data of the contract make sure to use the DoNotMerge setting. In gerenal, those that which to test through several contracts (let's say over a year period) will set the Merge Policy to MergeBackAdjusted or MergeNonBackAdjusted to have a continuous contract.

    For information on the Merge Policies please visit the following link: http://www.ninjatrader.com/support/h...rical_data.htm

    Please let me know if I may be of further assistance.

    Comment


      #3
      Patrick
      Does it mean that it will not affect the accuracy of my back testing regardless of the merge policy I use?

      Comment


        #4
        Hello sys_trader,

        Thank you for your response.

        This is going to depend on how you view the accuracy of your test. Do you need to test large ranges of dates? Or do you need that raw data tested for each contract? The raw data (DoNotMerge) would have no adjustments made to it, so it would be the truest form of the data for that contract.

        Comment


          #5
          Patrick
          I use the largest possible data set that Kinetic gives me to back test my system's robustness. I think I will use MergeNonBackAdjusted which does not apply the off-set value during a rollover. so all historical data are raw. That is good enough for me.

          Comment


            #6
            I understand these 3 merge policies are for futures contracts only. So they are ingnored when applied to stocks or currencies theoretically. However, my observations are different. My back testing results are different for the merge polices when applied to currencies. The results should be all the same, right?

            Comment


              #7
              That understanding is correct sys_trader, you refer to spot currencies though? So not the currency futures on Globex like 6E for example?
              BertrandNinjaTrader Customer Service

              Comment


                #8
                Originally posted by NinjaTrader_Bertrand View Post
                That understanding is correct sys_trader, you refer to spot currencies though? So not the currency futures on Globex like 6E for example?
                Correct like $EURUSD. I think this might be a data issue or a minor bug in Forward Walk Testing. But I found it does not affect the overall accuracy of my system. cool:

                Comment

                Latest Posts

                Collapse

                Topics Statistics Last Post
                Started by Shansen, 08-30-2019, 10:18 PM
                24 responses
                938 views
                0 likes
                Last Post spwizard  
                Started by Max238, Today, 01:28 AM
                0 responses
                3 views
                0 likes
                Last Post Max238
                by Max238
                 
                Started by rocketman7, Today, 01:00 AM
                0 responses
                1 view
                0 likes
                Last Post rocketman7  
                Started by wzgy0920, 04-20-2024, 06:09 PM
                2 responses
                27 views
                0 likes
                Last Post wzgy0920  
                Started by wzgy0920, 02-22-2024, 01:11 AM
                5 responses
                32 views
                0 likes
                Last Post wzgy0920  
                Working...
                X