Timeline for Percentile calculation for the combined group [duplicate]
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Dec 23, 2019 at 3:56 | history | suggested | Steffen Moritz | CC BY-SA 4.0 |
Removed additional comments
|
| Dec 23, 2019 at 0:47 | review | Suggested edits | |||
| S Dec 23, 2019 at 3:56 | |||||
| Sep 26, 2019 at 12:48 | history | closed |
Stephan Kolassa Peter Flom |
Duplicate of Should 99 percentile of a union of two sets be the average of 99 percentile of each set? | |
| Sep 26, 2019 at 7:40 | review | Close votes | |||
| Sep 26, 2019 at 12:48 | |||||
| Aug 12, 2019 at 18:02 | history | bumped | CommunityBot | This question has answers that may be good or bad; the system has marked it active so that they can be reviewed. | |
| Jul 11, 2019 at 12:40 | answer | added | Cadoiz | timeline score: 0 | |
| Mar 17, 2012 at 3:09 | comment | added | Cyan | To follow up on whuber's comment, if you can give more information about where these groups come from and/or what kinds of structure you expect to see in the data (and especially their relative sizes!) then it might be possible to work something out... | |
| Mar 16, 2012 at 20:48 | comment | added | whuber♦ | You can say very little when the groups have different sizes. E.g., let one group be enormous and contain 89.9% positive numbers and 10.1% 0's, so 0 is the tenth percentile. If the other groups all have positive values, then the combined 10th percentile is some unknown positive number. If the other groups all have negative values and contain more than 10% of all the data, the combined 10th percentile is some negative value: all you know is that it exceeds the most negative of the individual 10th percentiles. | |
| Mar 16, 2012 at 20:15 | history | asked | Saptarshi | CC BY-SA 3.0 |