Pages

Wednesday, July 14, 2021

economics, statistics. and inflation

Reminder: Please continue to e-mail subscribe to Salil's Statistical Ideas.  While Google terminates some of these site functions in July 2021, we will resume activities in some other format.  Also if you like, please purchase or review my books: Statistics Topics, or Colors and Numbers.


Econometricians prefer to have normalized inflation models that are magnetized at the long-term 2%.  As a result it's easier to rationalize away any aberrations that we see.  And this is "fine", until it isn't.  Statistical modelers prefer to model the actual non-parametric data, where it's evident that the rate of change continues to increase away from 2% at an alarming rate.  



It's not a slam dunk the economists will be right*.  I hope they are.  But to the degree monthly inflation continues to surprise the "experts", and to the degree that the price increase is spreading away from food and energy related items [or conversely there is a decrease in the number of deflationary items], the probability increases DRAMATICALLY that we will settle-up with lingering, non-transitory inflation many σ higher than a hypothesized 2% average.  

How much longer would you roll the dice speculating that prices will slow [dubious commentary no longer dispassionately grounded in the data]?  We are continuously increasing the probability that an uncomfortable recognition of this very different but highly plausible alternate course [not one symmetrically near 2%] would occur too late for policy makers to then counter.


* One of those top voices is now a new follower of this site:



No comments:

Post a Comment