MathGroup Archive 2000

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: continuously updating mean confidence interval (MeanCI)


Daniel Reeves wrote:
> 
> Suppose you're running a long simulation to generate a bunch of data
> points, and you want to compute the mean.  You can do this by just storing
> the cumulative sum and the count as you generate the data.
> 
> If you also want a 95% confidence interval on the mean, what is the
> minimum amount of state you need to store?
> 
> I think this reduces to finding the Variance, and I fear that the answer
> is "the whole history of data points"

No reason to fear. The sample variance is the the mean of the squares
minus the square of the mean, so the only extra thing you need is the
sum of the squares of your data points, for which you need not keep the
history.

-- 
John Doty		"You can't confuse me, that's my job."
Home: jpd at w-d.org
Work: jpd at space.mit.edu


  • Prev by Date: Industrial Electromagnetism
  • Next by Date: Simplification shortcomings?
  • Previous by thread: continuously updating mean confidence interval (MeanCI)
  • Next by thread: Any ideas?