Timeline for Calculate average of a set numbers with reported standard errors
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 22, 2019 at 21:56 | history | protected | Sycorax♦ | ||
| Feb 22, 2019 at 21:53 | comment | added | June Skeeter | I think you are looking for pooled variance. | |
| Mar 26, 2018 at 18:21 | answer | added | David Moles | timeline score: 2 | |
| Sep 25, 2015 at 9:39 | comment | added | Glen_b | The way the standard error of the sum (and hence the average) works depends on the assumptions made and how the predictions are generated (which will impact the correlation between them) | |
| S Sep 12, 2013 at 10:05 | history | suggested | Comp_Warrior | CC BY-SA 3.0 |
Slight editing; added references and error-propagation tag
|
| Sep 12, 2013 at 9:45 | review | Suggested edits | |||
| S Sep 12, 2013 at 10:05 | |||||
| Aug 10, 2012 at 6:24 | comment | added | chl | ../.. As you can see from the data, the first measurement (if assuming a normal distribution) would have Upper 95% of 45.7 and a lower 95% at 4. So when I average 365 numbers, what is the CI of all of them ? I did find this resource that talks about error propagation: Error Propagation but am not quite sure still. – DBirdmanAR | |
| Jan 17, 2012 at 1:25 | answer | added | Peter Ellis | timeline score: 6 | |
| Jan 16, 2012 at 14:45 | comment | added | user8559 | MånsT- Sorry, I've not tested it and realized that as well. It would likely be a log-normal with a very high peak near the Y axis and a long tail. Onur - the "Practical Example" is not relevant as it is a standard example of working from a known distribution with a known SD. In my case, each measurement has its own SE associated with it and its own Confidence Interval. What I actually want to do with my 365 numbers is say this: At a 95% Confidence Interval, the mean is above a certain standard, say 35. ../.. | |
| Jan 15, 2012 at 20:03 | comment | added | onur güngör | This gives a relatively detailed explanation of what confidence intervals are, in which conditions you can rely on them etc. In the "Practical Example" section, you will find a very similar example. | |
| Jan 15, 2012 at 17:18 | comment | added | DQdlM | An average is just a the sum of each item times its proportion. In the case of a normal average these would just be equal for each item (summing to 1 of course). So Is it appropriate to just use normal addition error propagation after multiplying by the proportion? | |
| Jan 15, 2012 at 9:11 | comment | added | MånsT | The Poisson distribution is used for discrete data whereas your data seems to be continuous. What I would like to know is how the standard errors were obtained. Are they related to the measrements themselves or were they somehow obtained separately? | |
| Jan 15, 2012 at 5:03 | history | migrated | from stackoverflow.com (revisions) | ||
| Jan 14, 2012 at 5:15 | comment | added | user918967 | I do not. For sake of argument we can say it is but it is likely Poisson because much of the other data I work with usually is. | |
| Jan 13, 2012 at 22:06 | comment | added | ahoffer | Do you know if the data normally distributed? | |
| Jan 13, 2012 at 21:00 | history | asked | user918967 | CC BY-SA 3.0 |