Announcement

Collapse
No announcement yet.

Cross correlation

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Cross correlation

    Hi,

    Does anyone know of a good (i.e. simple as possible) tutorial or resource for cross correlation? I am in the process of running a series of experiments with different equipment and as such would like to compare different filtering techniques. Ideally I would like to find, or create, an Excel spreadsheet to calculate cross correlations comparing raw data with the output from different filters.

    I have found scripts for R and Matlab but as I do not know how to use these software packages it doesn't help me learn the process. I would like to understand the whole process before I rely on it to draw conclusions.

    If anyone can point me in the direction of an Excel spreadsheet or a good tutorial that would help me create a spreadsheet of my own it would be much appreciated.

    Thanks
    Dan

  • #2
    Re: Cross correlation

    Hi Dan,

    Both SPSS and Minitab have a facility for calculating cross-correlation coefficients with or without phase lags.

    I have been looking into cross-correlations and their use quite extensively recently and found the literature to be quite ambiguous and contradictory at times. I found the following references to be of use, however:

    Derrick, T.R. & Thomas, J.M. (2004). Time series analysis: The cross-correlation function. In Innovative Analyses of Human Movement: Analytical Tools for Human Movement Research (edited by N. Stergiou), pp. 189-205. Champaign, IL: Human Kinetics.

    Winter, D.A. (2009). Biomechanics and Motor Control of Human Movement (4th Edition). Hoboken, NJ: Wiley. (Chapter 2)

    Cheers,

    Paul

    Comment


    • #3
      Re: Cross correlation

      Hi,

      R is very helpful and is very easy to use. Check this out: https://stat.ethz.ch/pipermail/r-hel...er/112363.html
      Hope it helps

      Comment


      • #4
        Re: Cross correlation

        Hi Paul,

        I have looked at Winters book and it’s very helpful, I was originally using the previous edition of that book which doesn't have the relevant section (guess I should upgrade!). I'll check the library for the other book too thanks!

        Dan

        Comment


        • #5
          Re: Cross correlation

          Originally posted by icegirlclaire View Post
          Hi,

          R is very helpful and is very easy to use. Check this out: https://stat.ethz.ch/pipermail/r-hel...er/112363.html
          Hope it helps
          I have heard good things about R, I have been dubious about learning R though as I am struggling to learn Matlab at the moment. Perhaps I should make the effort to learn both, however when I am learning a new theory I often come back to excel as I know it better than other software.
          Thanks for the heads up though.
          Dan

          Comment


          • #6
            Re: Cross correlation

            Originally posted by drobbins99 View Post
            Hi Paul,

            I have looked at Winters book and it’s very helpful, I was originally using the previous edition of that book which doesn't have the relevant section (guess I should upgrade!). I'll check the library for the other book too thanks!

            Dan
            Hi Dan,

            R is very powerful in statistics. But Matlab is more useful in all other kinds of computations. Good Luck!

            Comment


            • #7
              Re: Cross correlation

              Hi Dan,

              Yes, the section of Winter (2009) I am referring contains information that wasn't presented in previous versions of this book. Most of the salient information on CCs is available through Google books (http://books.google.co.uk/books?id=_...page&q&f=false). Section 2.1.5 provides information I couldn't find elsewhere, which would be very important if you try to do the calculations yourself.

              Paul

              Comment

              Working...
              X