|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Finding breakpoint in a piecewise model |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
Hello, I'd like to run a piecewise cohort-sequential model. Two questions: 1. I haven't seen examples of piecewise cohort sequential models in the literature. In principle, is there any reason why one can't use the piecewise approach within a cohort-sequential model? 2. The model spans 43 years, so it's not obvious where the breakpoint should be. I could simply plot the data and "eyeball" it, but, is there a more precise way to determine the optimal location for a breakpoint? Thanks, Jordan |
|
|
1. No. 2. I would look at the curves to see where the break may be. The is a literature on change point analysis that you might want to look at. Mplus does not yet do this. |
|
Back to top |
|
|