Timeline for Scalar-on-function regression with random initial time
Current License: CC BY-SA 3.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 1, 2017 at 9:59 | comment | added | DeltaIV | @chRrr thanks for the pointer to the paper - it's massive! I'll try to read it this week. I'm not sure that this simply a sparse data problem though - it's not only that the time series are irregularly sampled, it's also that they have widely different durations. If you have a look at the plot on the right, you can clearly see that, for some subjects, the length of the time series is just one third (or less) of the length for other subjects. | |
| Jul 31, 2017 at 13:55 | comment | added | chRrr | your functional linear regression model seems fine for me. your additional problem with the irregular design points (assuming all points are theoretically in a common intervall) may also be known as sparseness in this context. You may want to take a look at "Functional Data Analysis for Sparse Longitudinal Data" by Yao, Müller and Wang. | |
| Jul 31, 2017 at 13:12 | history | asked | DeltaIV | CC BY-SA 3.0 |
