Timeline for Average curve from set of curves
Current License: CC BY-SA 3.0
18 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 16, 2016 at 18:44 | vote | accept | Arkoudinos | ||
| Feb 16, 2016 at 15:32 | history | tweeted | twitter.com/StackStats/status/699617279886159872 | ||
| Feb 14, 2016 at 0:45 | answer | added | usεr11852 | timeline score: 1 | |
| Feb 12, 2016 at 23:15 | history | edited | Arkoudinos | CC BY-SA 3.0 |
Extra information about the problem
|
| Feb 12, 2016 at 22:11 | comment | added | Arkoudinos | Let us continue this discussion in chat. | |
| Feb 12, 2016 at 21:57 | comment | added | usεr11852 | So you mean you do not have fixed measurements times? If that is the case I think you definitely need to use either an LME or a functional data analysis technique instead of a standard longitudinal data analysis as I suggested earlier. Is that sample from one patient or is it a subsample from one patient? | |
| Feb 12, 2016 at 21:51 | comment | added | Arkoudinos | @usεr11852 Of course. I have 4 datasets that are consisting from about 15 subjects each (about 60 persons in total). Each dataset includes a time series for each patient and in this time-series the stride interval is being measured. Since this interval is not regular, the measurements aren't regular either. So, to have a picture the time-series looks like this: bpaste.net/show/8aaffb8d1432 | |
| Feb 12, 2016 at 21:32 | comment | added | usεr11852 |
@Arkoudinos: Cool, thanks for clarifying both points; the function fitlme is the one you want to use in this case. Can you please describe your dataset in a bit more detail? In that way we will be able to help you more substantially.
|
|
| Feb 12, 2016 at 21:30 | comment | added | Arkoudinos | @gung Unfortunately, I don't have the slightest idea about mixed-effects model... I'm going to take a look at it though. | |
| Feb 12, 2016 at 21:27 | comment | added | Arkoudinos | @usεr11852 I do not have two measurements per subject, but a time series which includes the specific time an event happened and the duration of that event (for example, a patient's foot "freeze" for 300ms). I'm using R2014a and yes I'm having access to the MATLAB Statistics Toolbox. Do you have any idea? | |
| Feb 12, 2016 at 18:52 | answer | added | America | timeline score: 0 | |
| Feb 12, 2016 at 18:45 | comment | added | usεr11852 | @Arkoudinos: In general, try to get a hold of the book "Statistical Methods for the Analysis of Repeated Measurements" by C.S.Davis or a book of similar applied longitudinal analysis scope. I think it will be a very valuable resource for you in this line of work. | |
| Feb 12, 2016 at 18:37 | comment | added | gung - Reinstate Monica | @Aksakal, mixed-effects model & multi-level models are ultimately equivalent. They have different terminology, etc., for historical reasons. | |
| Feb 12, 2016 at 18:35 | comment | added | usεr11852 | Arkoudinos, do you have only two measurements per subject or it just happens that all your "curves" are straight lines? Also do you have access to the MATLAB Statistics ToolBox and which MATLAB version are you using? (MATLAB really upgraded its Stats capabilities lately.) @Aksakal: I think that is gung's idea too. | |
| Feb 12, 2016 at 18:33 | comment | added | Aksakal | @gung, possibly multi-level model | |
| Feb 12, 2016 at 18:30 | comment | added | gung - Reinstate Monica | You should probably use a mixed-effects model from the start. | |
| Feb 12, 2016 at 17:59 | review | First posts | |||
| Feb 12, 2016 at 18:30 | |||||
| Feb 12, 2016 at 17:50 | history | asked | Arkoudinos | CC BY-SA 3.0 |