Options

What is Needed to Make this Random Forest Model Predict into the Future?

SkyTraderSkyTrader Member Posts: 88 Contributor II
in Help
Hi, 

I've used the Apply Model operator as suggested to try and make this Random Forest model to predict the "future" -- but it still doesn't predict beyond the last data of my data set (15th Sept 2020)?

I've attached the process to see if anyone can figure out what is needed to get it to make forward future predictions?





Any suggestion much appreciated,

Many thanks.
Tagged:

Answers

  • Options
    MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,507 RM Data Scientist
    Hi,
    the easiest way would be the lag operator and just shift the label attribute.
    Best,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • Options
    SkyTraderSkyTrader Member Posts: 88 Contributor II
    Thanks Martin, @mschmitz
    I've had a lot of problems despite re-importing my Excel data using the wizard many times. The attribute is not found anymore (I also had the issue of my Close attribute not being found because of an errant * in the Excel file). Not sure how to proceed as this import / attribute issue has been ongoing for a month now - it used to work when I first started using RM.

    So the end result is I get this:







    I'm assuming I'm using the lag operator correctly? I also tried ticking "include special attributes," 

    Cheers,
  • Options
    hughesfleming68hughesfleming68 Member Posts: 323 Unicorn
    Hi SkyTrader, I would use also experiment with the windowing operator and sliding window validation. The results of your testing can change dramatically depending on your split point. 

    regards,

    Alex
Sign In or Register to comment.