Friday, April 1, 2016

Growth state diagrams

I wanted to show the plausibility of these "growth state diagrams" (here, here, here or here):


... using real data. Therefore I grabbed some CPI component data (as a proxy for growth) and plotted the histogram of CPI component inflation rates over time (past 10 years) in an animation (data from here):


1 comment:

  1. Very nice, Jason. I've wondered about the possibility of classifying different kinds of inflation regimes by the distribution of the members in the CPI basket. Would be interesting to dig this data up going forward from the 1950s to really span a broad set of macro regimes. One thing I've constantly found frustrating in finance / trading is the desire to express entire distributions as one number (avg, median), or the tendency to think that even sophisticated fits (copula modeling, etc.) adequately capture the behavior of things. I'm much more a fan of trying to find intelligent ways to use non-parametric distribution measures.

    ReplyDelete

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.