Timeline for Checking for overdispersion in a Poisson model
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 27, 2023 at 18:20 | comment | added | Jason Bohenek | Recommenting on this because my previous (now deleted) answer was garbage. So yes, >1 = overdispersed, <1 = underdispersed. Poisson distribution is defined by having one parameter that describes both the mean and variance. Thus, mean and variance must be equivalent, which of course is rarely true for any real data. If the ratio is >1, then it means the data are overdispersed and the variance is greater than the mean. If the ratio is <1, then it means the data are underdispersed, and the variance is less than the mean. Is the ratio = 1, then it means the mean = variance. | |
| Aug 13, 2020 at 0:07 | comment | added | wolfsatthedoor | Why should it be near one? | |
| May 1, 2018 at 18:22 | history | edited | gung - Reinstate Monica | CC BY-SA 3.0 |
formatted
|
| May 1, 2018 at 18:19 | history | edited | Jason Bohenek | CC BY-SA 3.0 |
added 13 characters in body
|
| Apr 3, 2018 at 18:08 | review | Late answers | |||
| Apr 3, 2018 at 18:29 | |||||
| Apr 3, 2018 at 17:53 | review | First posts | |||
| Apr 3, 2018 at 20:58 | |||||
| Apr 3, 2018 at 17:51 | history | answered | Jason Bohenek | CC BY-SA 3.0 |