Home Forums Forecasting with Deseasonalized data

  • This topic has 1 reply, 2 voices, and was last updated 7 years ago by hessinemaaoui.
Viewing 1 post (of 1 total)
  • Author
    Posts
  • #2354
    hessinemaaoui
    Keymaster

    Hi,
    Not sure if this is the right forum but here is my question:

    I am new to using Suanshu .NET. and I want to integrate it into my trading system OpenQuant. I was hoping you could give me an idea of how to perform a couple of tasks to get me going? For example, how do I:
    1. “Deseasonalize” hourly volume data (I don’t actually use hourly but much shorter time frames but this makes things simpler) in the futures market (since volume is heavily related to time of day) using a simple ratio to moving average method or any other method for that matter.
    2. Use some form of trend identification method, for example ordinary least squares (methodology is more important to me over specific method) and add back the seasonal component to forecast the expected volume going forward for the next bar volume.

Viewing 1 post (of 1 total)
  • You must be logged in to reply to this topic.