Missing data on technical analysis

Discussion in 'Data Sets and Feeds' started by Maria, Nov 6, 2006.

  1. Maria



    Can anyone tell me how should I proceed when I have data from a gas future like Winter 07 and want to analyze it since 2003? The problem is that in 2003, 2004 there was not much trading of this product so there is lots of "holes" in the data. Should I use it anyway? Should I fill in the holes with 0 quantities and last closing prices, is there a standard procedure for this case or you just should not do technical analysis when there is lots of missing data?
    Thank you for the help.

  2. Emrosie


    My question would be what time of analysis are you going to attempt, spreads? winter strip prices? Feb outright prices?

    If you're looking to analyze the winter time frame that we are approaching you can chose what that same winter spread did during the month of November. For example, what was the shape of the winter months of '06 during November of '05, the winter months of '05 in the month of November '04, etc.

    This gives you a better objective risk analysis, just as you said, because there isn't enough action in those months when they first come on the board.
  3. Similar to what was said previously, use the more current contract months and create your own "continuation" data.
  4. Maria


    Thank you for your replies. I guess it is better then filling in the "holes" in the data. I will try that.