Re: standard deviation


I'm curious what you guys are actually using this information for.

The idea of computing a standard deviation of what is effectively a weighted running average seems like a strange concept to me.

jeff

Najaf A. Shah wrote:
Hi Fernando,
  I'm doing the same for my tests.  You can simply import the data
into Excel and use the STDEV function on the list of values.  I'm
guessing you  want to do it on transfer speed.  One thing to keep in
mind is that the 1 second interval might be a little low; on a wired
connection it might be okay but on a wireless or otherwise noisy
connection you might get really low values the first second and really
high values the second (thus artificially increasing your deviation).

Hope this helps.



On 2/6/06, Fernando Cerutti <facerutti --at-- gmail.com> wrote:

Hi all,
I would like to estimate the standard deviation of iperf jitter measures for
udp flows
How can I do this with the measures already stored in files reported with 1
second interval (-i 1)
Thanks for advances,

Fernando Cerutti
http://inf.unisul.br/~cerutti





Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index