I'm familiar with linear interpolation. I tried to search some details about piece-wise linear interpolation but failed to understand this method. Can someone explain the difference between these two interpolation methods or perhaps offer some references to learn?


piece-wise models, at least in my field, mean a regression model with where the slope is "broken" into different pieces or sections. This is commonly done in models dealing with repeated measures across time. For example, imagine that you have a time series model with 10 data points. You administered your treatment at data point 4. You would have two "sections" of your regression slope. One before treatment and one after. Comparing the two slopes lets you know things like the difference between the two treatments and the difference in rate of change.

My suggestion is read one of the very handy books on time series modeling. I have no suggestions since the resources I would suggest would assume a background in R which I do not know if you have.

  • $\begingroup$ Great answer, I appreciate the resources you would like to suggest because I do have a background in R. $\endgroup$ – YQ.Wang Jun 3 '17 at 13:03
  • $\begingroup$ I really think that Julian Faraway does a great job of mixing R and statistics well. Look for his book "linear models with R" and "extending the linear model with R". As for a book on time series that I like "Introduction to Time Series Analysis" by Mark Pickup. That's a little book that does a good job explaining. $\endgroup$ – JWH2006 Jun 3 '17 at 13:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.