On Mon, 30 Oct 2000 18:54:12 -0800, "G. Anthony Reina" <[EMAIL PROTECTED]>
wrote:
> I'm having a problem concerning cross-correlation and was hoping someone
> could help explain.
>
> Here's what I'm doing:
>
> I create two random signals (each 100 points from gaussian distribution
> from -1 to 1) and find the maximum cross-correlation value (either
> negative or positive, whichever has the larger absolute value). I'm
> repeating this 10,000 times in MATLAB so that I can get a feel for the
> distribution of maximum cross-correlation between the signals.
>
> The distribution that I get looks like two gaussian curves (bimodal) at
> -0.2 and 0.2. I know that for 100 observations a correlation is
> significant if > 0.2 (p ~ 0.05). However, I guess I'm expecting that the
> distribution would have been unimodal around zero with tails going to
> -0.2 and 0.2.
< snip, rest >
I don't read your MATLAB program, but it sounds as if
you are expecting the Max(50 or 100 continuous values)
to have a mode at 0.
For Max(positive X), should you not expect a 0 density at zero?
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
