Timeline for Biological time-series data with random variation: is regression suitable and centering variables to remove year-effect
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 7 at 4:02 | comment | added | EdM | @Meerkat then perhaps canopy volume is a mediator of the effects of management treatment? There is a literature and special methods devoted to mediators, which are both affected by treatments and whose treatment-associated changes account (in part, at least) for the effects of the treatment on the outcome of interest. Methods for analyzing mediation do go beyond standard regression, although regression tools are involved. | |
| Nov 6 at 23:51 | comment | added | Meerkat | One quick comment before I re-read and re-re-read your answer to properly understand it. My statistician said that my explanatory variable (canopy volume) is not an observation such as humidity, because it IS controlled to a certain extent - by the management treatments. So it falls between the cracks, because it can't be set to a desired value, but it can be controlled. Apparently, D&S state that “an assumption of regression analysis is that the predictor variables are NOT subject to random variation”, which in my case they are. | |
| Nov 6 at 21:14 | history | answered | EdM | CC BY-SA 4.0 |