Timeline for General Question: Should Legitimate Outliers in the Data be Included or Excluded from Statistical Models? [duplicate]
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 14, 2022 at 13:19 | history | closed |
mkt kjetil b halvorsen♦ regression Users with the regression badge or a synonym can single-handedly close regression questions as duplicates and reopen them as needed. |
Duplicate of Is it OK to remove outliers from data? | |
| Sep 13, 2022 at 18:42 | review | Close votes | |||
| Sep 14, 2022 at 13:19 | |||||
| Jan 6, 2022 at 2:55 | comment | added | stats_noob | Thanks everyone for your replies! | |
| Jan 2, 2022 at 22:52 | comment | added | BigBendRegion | If you are trying to estimate a data generating process that produces occasional extreme values, then the estimated model without outliers will obviously be worse in the sense that it underestimates process variability. | |
| Jan 2, 2022 at 8:09 | comment | added | BruceET | The answer hangs on your definition of 'legitimate'. Of course, obvious mistakes should be removed (773 year-old participants, negative blood glucose levels, verifiable data entry errors, etc.) But values that are simply unusual may contain useful information. | |
| Jan 2, 2022 at 4:27 | answer | added | Scar_Face | timeline score: 1 | |
| Jan 2, 2022 at 4:17 | history | edited | stats_noob | CC BY-SA 4.0 |
added 269 characters in body
|
| Jan 2, 2022 at 3:49 | history | asked | stats_noob | CC BY-SA 4.0 |