Skip to content

Splitting time series into blocks/regimes?

3 messages · Michael, R. Michael Weylandt, Whit Armstrong

#
I find this idea methodologically iffy, but perhaps you could do a
little bit to make it a well posed question:

i) Are you picking these periods as relatively equal divisions or time
or as ex-post artifacts of "all that mess"

ii) What is "the clustering algorithm"? Aren't there quite a few?

iii) What does this mean: "the 3 big blocks/regimes need to be
contiguous within each block/regime itself..."

Perhaps you should look into reformulating your idea in terms of
regime switching models.

Michael Weylandt
On Tue, Feb 28, 2012 at 12:55 PM, Michael <comtech.usa at gmail.com> wrote:
#
It's not easy to do this.

http://dss.ucsd.edu/~jhamilto/palgrav1.pdf
http://www.econ.washington.edu/user/cnelson/markov/prgmlist.htm

Most people use gibbs sampling or other mcmc methods.

as it happens, I have some software you can use if you don't mind
getting your hands dirty.
https://github.com/armstrtw/CppBugs

or you can just use BUGS/JAGS.

also, some code is here:
http://faculty.chicagobooth.edu/hedibert.lopes/teaching/sa2011/sa2011.html

-Whit



On Wed, Feb 29, 2012 at 7:24 AM, R. Michael Weylandt
<michael.weylandt at gmail.com> wrote: