+ Reply to Thread
Results 1 to 2 of 2

Thread: Time Series Regression Approach

  1. #1

    Time Series Regression Approach




    My goal is to predict monthly S&P 500 values using a single indicator data set (For example, Unemployment Rates )

    My approach has been to extract lots of features using lag values, then fit the model using the best set of features. I am using Python and SciKit learn, and fitting using a basic linear model.

    y: S&P 500
    X: UR, UR-1, UR-2, ...

    (Where UR = Unemployment Rate for that month, UR - 1= previous months, etc)

    Questions:

    Is the general approach way off? Im sure this problem has been many times, should I be thinking about it completely different?

    Feature Reduction/Selection - I suspect there is strong autocorrelation with these indicators, and plan on using PCA to address. Any concerns with that? What is the best method in this context?

  2. #2
    Human
    Points: 12,676, Level: 73
    Level completed: 57%, Points required for next Level: 174
    Awards:
    Master Tagger
    GretaGarbo's Avatar
    Posts
    1,362
    Thanks
    455
    Thanked 462 Times in 402 Posts

    Re: Time Series Regression Approach


    Sorry, buy I don't think that it is possible to predict speculative markets. But I think that there are many other areas where the methods can be useful.

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats