Passing matrix or vector objects to gsl_statistic routines.

[Mike Benton]

Hi,

Is it possible to use a matrix or vector object, say gsl_matrix *dta or
gsl_vector *vdta, and pass it to any of the gsl_statistic routines.

For example:

dta is populated with data.

double get_skew (int variable) {
    double skew;
    skew = gsl_stats_skew (dta[variable], 1, number_of_cases);
    return skew;
}

or

void get_regression (int x, int y) {

    double c0, c1, cov00, cov01, cov11, chisq;

    gsl_fit_linear (dta[x], 1, dta[y], 1, number_of_cases,
                   &c0, &c1, &cov00, &cov01, &cov11,
                   &chisq);

   printf ("# best fit: Y = %g + %g X\n", c0, c1);
   printf ("# covariance matrix:\n");
   printf ("# [ %g, %g\n#   %g, %g]\n",
          cov00, cov01, cov01, cov11);
   printf ("# chisq = %g\n", chisq);

}

I have been trying and have had no success.  My desire is to pass the object
to the routine without creating an array.

I am programming in C on Linux, using version 1.4 of GSL.  Lastly, I am
re-new at c (I stopped programming in 92 and I have just started programming
again.).

Thanks for you help,

Mike H. Benton