Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Correlation Coefficient problem

Status
Not open for further replies.

abhigopal

Member level 3
Joined
Dec 21, 2004
Messages
61
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
624
related:www.rocksolidimages.com/pdf/semblance.pdf

Hi all
My problem is kinda weird. I have two arrays of numbers and I want a pairwise correlation between these two numbers. I use MATLAB 7.0's inbuilt function namely corr.m .. My problem is tht one of these arrays has only real numbers and the other one has only complex numbers. Both these arrays are of equal length.

When I compute the correlation, I get a complex number as my answer. I take the absolute value of the answer and My answer turns out to be greater than 1. But the correlation coeff always has a value between -1 and +1. I also typed out a code for computing the correlation coefficient and ended up with the same result.

Does anyone have any explanations on why this is happening.??
Please help

Thanks in advance
Abhishek

P.S- Moderators, I could not find an appropriate forum for postin this message. If required, please move it to an appropriate forum.. Thanks
 

The coefficient may be the complex number you got out of the program. Are the real and the imaginary parts separately within -1 to +1?
 

Flatulent,

No, the real and imaginary parts are not lying between 1 and -1. One of them is greater than 1 at times. And if the complex number I got is actually the coefficient, then how would I know the extent upto which the two arrays are correlated.
And, this does not occur in all cases where one array is purely real and another is purely imaginary. Just in some cases..
 

**broken link removed** is an interesting discussion of what you are doing. The real and imaginary parts of the coefficient measure different parameters of the correlation.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top