This repository was archived by the owner on Nov 13, 2021. It is now read-only.
This repository was archived by the owner on Nov 13, 2021. It is now read-only.
Open
Description
I've been going through the code for edm-multi.cpp, and I some general confusion about the algorithm being carried out. I'll keep my questions limited at first so that perhaps the answers to them can alleviate my confusion on the rest.
- Is the statistic used in the algorithm to detect multiple breakouts (in edm-multi.cpp) to measure the significance of a breakout at a certain point the same as in that for detection of a at most one breakout (i.e. in edmTail.cpp)? I feel like I understand the process used in edmTail, especially thanks to the paper written introducing the breakout detection algorithm. But it looks like edm-multi measures something slightly different--as best as I can tell it's looking for a shift in the median with some sort of optional penalization.
- Is it possible to get some explanation as to what is going on in lines 73-78 of edm-multi.cpp? I think one of the main things confusing me about this is the
F[t]term in the definition of tmp. It seems like this would have an unwanted accumulative effect on the statistic. - Am I correct in saying that when
*G = Quadratic, breakouts that occur earlier in the time series are favored more than those that occur later? It seems like the more breakouts that have been observed, the more this term will penalize the tmp statistic.
Metadata
Metadata
Assignees
Labels
No labels