Timeline for How to fit the right cross-classified multilevel model
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 25, 2020 at 18:48 | history | edited | Robert Long | CC BY-SA 4.0 |
deleted 18 characters in body
|
| Aug 25, 2020 at 18:43 | comment | added | Robert Long | @SeydouGORO I have added a simulation to demonstrate that it doesn't matter what level a variable varies at. | |
| Aug 25, 2020 at 18:42 | history | edited | Robert Long | CC BY-SA 4.0 |
Added simulation for fixed effects levels
|
| Aug 25, 2020 at 18:14 | vote | accept | Seydou GORO | ||
| Aug 25, 2020 at 17:49 | comment | added | Robert Long | @SeydouGORO Perhaps it would be better to say that it does not need to know, because it has no impact on the estimation procedure. There is no way for you to tell the software at what level it varies anyway! | |
| Aug 25, 2020 at 16:51 | comment | added | Seydou GORO | Thank you for all your clarifications, which help me a lot. But how can the software know, for example, that the variables HHI and hospital status are level 2 variables and that they are relative to the hospital. Also, that the deprivation index variable is related to the city and that they are not related to the patients, to avoid some bias. | |
| Aug 25, 2020 at 15:52 | history | answered | Robert Long | CC BY-SA 4.0 |