Least Squares with Breakpoints

For notifying us of what you believe are bugs or errors in EViews.
Please ensure your copy of EViews is up-to-date before posting.

Moderators: EViews Gareth, EViews Moderator

maxchen
Posts: 191
Joined: Fri Oct 10, 2008 4:03 pm

Least Squares with Breakpoints

Postby maxchen » Fri Jul 24, 2015 6:21 am

EV9 20150723 x86

Code: Select all

wfopen "C:\EViews9\Example Files\EV9 Manual Data\Chapter 31 - Least Squares with Breakpoints\hansen_jep.wf1"
smpl 1947M01 2001M04
frml y = D(DUR)*12*100
equation eq01.breakls(cov=white, nodf) y y(-1) c   'Breaks: 1965M08, 1991M06
equation eq02.breakls(cov=white, nodf, size=10) y y(-1) c   'Breaks: 1965M08, 1991M06
'UG2, p413
equation eq03.breakls(cov=white, nodf, trim=5, size=10) y y(-1) c   'Breaks: 1963M12, 1994M12
'


'UG2 p415 vs p413
freeze(tbbk) eq03.breakspec

discuss
1 eq03 vs eq02 (Trimming 0.15), big difference
2 UG2 p415, not match p413, the break points
3 what is "Repartition"

EViews Glenn
EViews Developer
Posts: 2671
Joined: Wed Oct 15, 2008 9:17 am

Re: Least Squares with Breakpoints

Postby EViews Glenn » Fri Jul 24, 2015 10:00 am

1. Yes there is a big difference, as the 1994m12 break is not admissible under 15 percent trimming, while it is under 5 percent trimming.

2. The EQ02 results shouldn't match the output from UG2, p. 413. The EQ03 results should match the UG2 example, which they do. Note that your EQ02 uses 15 percent trimming while EQ03 uses 5 percent trimming.

3. Repartitioning is a refinement of multiple breaks determined under sequential testing. From the EViews documentation:
Once the sequential breakpoints have been determined, Bai recommends a refinement procedure whereby breakpoints are re-estimated if they are obtained from a subsample containing more than one break. This procedure is required so that the breakpoint estimates have the same limiting distribution as those obtained from the global optimization procedure.

You can consult the original paper for further detail.

maxchen
Posts: 191
Joined: Fri Oct 10, 2008 4:03 pm

Re: Least Squares with Breakpoints

Postby maxchen » Fri Jul 24, 2015 5:38 pm

I see.

so the breakpoints are ugly in real data, they are for large number, an art for real data.

saadallah
Posts: 15
Joined: Wed Dec 07, 2016 4:59 am

Re: Least Squares with Breakpoints

Postby saadallah » Sun Dec 18, 2016 5:03 am

Hello,

Does anyone know how can i store the break date of a "least squares with breakpoints".

In fact i'm performing a least squares with a max one breakpoint : "equation eq1.breakls(maxbreaks=1) y c x" but i would like to know if i can store the breakdate and the 2 coefficients?

Thank you


Return to “Bug Reports”

Who is online

Users browsing this forum: No registered users and 17 guests