Timeline for AIC and BIC Manual Calculations Are a Bit Off From Statsmodels Estimates in Python
Current License: CC BY-SA 4.0
3 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 9, 2023 at 16:50 | comment | added | Josef |
There are differences across packages for example in whether the scale estimate is included (in R) or not included (in statsmodels default) in the parameter count k.
|
|
| May 9, 2023 at 16:49 | comment | added | Josef | The values are the same at print precision, 4 to 5 digits. To see a difference you need to print more decimals. Then the difference might be just numerical noise from the floating point computation (or other sources, e.g. convergence tolerance of parameter estimates) | |
| May 9, 2023 at 3:10 | history | asked | GSA | CC BY-SA 4.0 |