Wednesday, January 14, 2015

Better than Fed

Hopefully restore some of my credibility in the wake of messing up the Japan model, here's how the information transfer model is doing relative to the Federal Reserve Board's prediction. I added some annual averages as thick dotted curves -- however be aware the annual average doesn't include the final inflation number or GDP number from 2014 (so the average is subject to revision). Anyway, here is the inflation prediction:


And here is the RGDP prediction:


Note that the differences between the red solid curve and the thick dotted curves appears to be due to data revisions.
Posted by at
Labels:

4 comments:

  1. Todd ZorickJanuary 14, 2015 at 10:47 AM

    I appreciate posts like this, because you are making strong predictions which can be tested against the unknown future. If I were a betting man I would consider this a good time to take on interest-rate risk in the form of a floating interest rate mortgage, based upon ITE. We are planning on refinancing our mortgage this year, will have to strongly consider the advantages of the lower rate ARM products.

    BTW I should mention that after reading your inspirational article from Fielitz and Borchardt, 2009, I have been playing around with the application of these ITM techniques to brain dynamical systems I am studying, with some potentially very interesting results. However, I would like a consultation about my methods and how best to apply them. Are you interested in hearing what I have done and working through an application to neuroscience with me? Off the blog, of course.

    ReplyDelete
    Replies
    1. Jason SmithJanuary 14, 2015 at 11:18 AM

      Cheers, Todd. Although I would caution against using just the ITM to make investment decisions! :)

      Although lots of people are predicting low interest rates for a long time -- that's the basis behind the whole 'secular stagnation' argument from Larry Summers.

      And certainly -- I'd love to hear about it. Although my background is in physics, so I'll probably have some dumb neuroscience questions. There is an email box on the side of this blog that sends a message to my gmail account without posting anything to the blog. That was how Fielitz and Borchardt originally got in touch with me.

      Delete
    2. Peter and GuenterJanuary 15, 2015 at 8:01 AM

      Dear Todd,
      We regularly read Jason’s blog and are glad to read that you see a potential application of the information transfer concept to neuroscience. This would be a very interesting application of the concept to non-physical (biological) systems/processes. Please let us know (you can contact us by the email addresses in the article) if you have any (published) results.
      Best regards
      Peter and Guenter

      Delete
    3. Todd ZorickFebruary 9, 2015 at 8:37 AM

      Thanks to everyone for the offer, it is much appreciated. Sorry to be so long in responding, but I posted this shortly before I left for vacation, which was in southern Baja Mexico, and the communications infrastructure had been damaged by a hurricane late last year, and yet to be repaired. Just getting back to the real world now and will send you all a precis of the idea I have in mind.

      Delete

Comments are welcome. Please see the Moderation and comment policy.

Also, try to avoid the use of dollar signs as they interfere with my setup of mathjax. I left it set up that way because I think this is funny for an economics blog. You can use € or £ instead.

Note: Only a member of this blog may post a comment.

Subscribe to: Post Comments (Atom)