Timeline for Median from two unmergable datasets
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 12, 2020 at 18:01 | vote | accept | Arktik | ||
| Nov 12, 2020 at 15:23 | answer | added | whuber♦ | timeline score: 4 | |
| Nov 11, 2020 at 23:20 | comment | added | Arktik | OK, thanks! I think mixture is the keyword here. | |
| Nov 11, 2020 at 23:10 | comment | added | whuber♦ | By means of a range of quantiles (along with the dataset sizes) you can closely approximate the distribution in each system; the combined dataset is the mixture of the two, whose quantiles are easily calculated. As an example of how you could go about choosing suitable quantiles, see stats.stackexchange.com/a/35268/919 for one proposal that provides good control over the precision. | |
| Nov 11, 2020 at 22:55 | history | edited | Arktik | CC BY-SA 4.0 |
added 1 character in body
|
| Nov 11, 2020 at 22:54 | comment | added | Arktik | Thanks whuber. I was struggling to come up with search terms for this. I will have to take a good look at those links. Do you think the sampling approach from quantiles is too naive? Regarding more information: a) Doesn't have to be very accurate, it's just for summary tables for a paper, b) I can carry over non-sample-level information (subject to statistical disclosure control), e.g. summaries like quantiles that were calculated with more than 10 samples, which is fine as N is large, c) I have multiple datasets not just two but I think I can automate the iterations -- not tried yet. | |
| Nov 11, 2020 at 22:39 | comment | added | whuber♦ | Thank you. See this search. stats.stackexchange.com/questions/7959 is directly applicable. It would help to know (a) how accurately you need to compute the median, (b) what information you can carry over from one system to the other, and (c) how often you can iterate back and forth between systems. | |
| Nov 11, 2020 at 22:35 | comment | added | Arktik | I've edited the questions. Thanks. | |
| Nov 11, 2020 at 22:35 | history | edited | Arktik | CC BY-SA 4.0 |
added 135 characters in body
|
| Nov 11, 2020 at 20:04 | comment | added | whuber♦ | Could you explain what you mean by "can't bring the raw data together"? | |
| Nov 11, 2020 at 19:57 | review | First posts | |||
| Nov 11, 2020 at 22:06 | |||||
| Nov 11, 2020 at 19:55 | history | asked | Arktik | CC BY-SA 4.0 |