Re: compare 2 samples
As Dave suggested, your data are highly non-normal. For non-negative times, the largest standard deviation relative to the mean occurs in the following extreme case.
N-1 measurements are zero and a single measurement is c (a constant). The mean is c/N and the standard deviation is c/sqrt(N). The ratio of the standard deviation to the mean is sqrt(N).
I suspect that in your data, the preponderance of the measurements are close to zero, with a very few being moderate or large.
Jack
.
Relevant Pages
- Re: Demonstration that least squares give maximum likelihood
... Call the measurements m, m, ..., m. ... mean m and standard deviation sigma. ... It is generally accepted in our field that you can usually assume normality, partly because the errors have multiple causes. ... Another responder alluded to concerns that - outliers apart - there are suspicions that the tips of the tails of real analytical data distributions are bigger and longer than the tails of the normal distribution. ...
(sci.stat.edu) - Re: compare 2 samples
... The ratio of the standard deviation to the mean is sqrt. ... I suspect that in your data, the preponderance of the measurements are close to zero, with a very few being moderate or large. ... Changing the logging code is easy, but if I want to compress the release cycle on the software, politically I'll have to add some high value features first... ...
(sci.stat.math) - Re: compare 2 samples
... The mean is c/N and the standard deviation is c/sqrt. ... I suspect that in your data, the preponderance of the measurements ... are close to zero, with a very few being moderate or large. ... I'll never know unless I rework the logging code to dump out the raw ...
(sci.stat.math) - Re: compare 2 samples
... The ratio of the standard deviation to the mean is sqrt. ... I suspect that in your data, the preponderance of the measurements are close to zero, with a very few being moderate or large. ... "longest delays" is what matters - using some cutoff, ... the delays seem to differ, it is quite possible that ...
(sci.stat.math) - Re: Population SD or sample SD with many replicates?
... indefinite because there is an indefinite number of possible ... to be exact the standard deviation would always have ... sample sets (usually more than 20 measurements). ... practice to use the population SD formula for measurements with many ...
(sci.chem.analytical) |
|
