diff options
Diffstat (limited to 'math/curfit/doc/cvepower.hlp')
| -rw-r--r-- | math/curfit/doc/cvepower.hlp | 55 |
1 files changed, 55 insertions, 0 deletions
diff --git a/math/curfit/doc/cvepower.hlp b/math/curfit/doc/cvepower.hlp new file mode 100644 index 00000000..58e78dae --- /dev/null +++ b/math/curfit/doc/cvepower.hlp @@ -0,0 +1,55 @@ +.help cvepower Jun95 "Curfit Package" +.ih +NAME +cvepower -- compute the errors of the equivalent power series +.ih +SYNOPSIS +cvepower (cv, y, weight, yfit, npts, chisqr, errors) + +.nf +pointer cv # curve descriptor +real y[] # array of y data points +weight weight[] # array of weights +real yfit[] # array of fitted data points +int npts # number of points +real chisqr # the standard deviation of the fit +real errors[] # standard deviations of the power series coefficients +.fi +.ih +ARGUMENTS +.ls cv +Pointer to the curve descriptor structure +.le +.ls y +Array of y data points +.le +.ls yfit +Array of fitted y values +.le +.ls npts +The number of points +.le +.ls chisqr +Reduced chi-squared of the fit. +.le +.ls errors +Array of standard deviations of the equivalent power series coefficients. +.le +.ih +DESCRIPTION +Calculate the reduced chi-squared of the fit and the standard deviation +of the equivalent power series coefficients for fitted Legendre and +Chebyshev polynomials. The errors are rescaled to the equivalent power +series and to the original data range. +.ih +NOTES +The standard deviation of the fit is the square root of the sum of the +weighted squares of the residuals divided by the number of degrees of freedom. +If the weights are equal, then the reduced chi-squared is the +variance of the fit +The error of the j-th coefficient is the square root of the j-th diagonal +element of the inverse of the data matrix. If the weights are equal to one, +then the errors are scaled by the square root of the variance of the data. +.ih +SEE ALSO +.endhelp |