Next: FREQUENCY Up: COMPARE Prev: Parameters
Method
Chi-squared is calculated as Sum[(data-model)**2/variance], where
variance is the variance of the data, unless the model array also
has a VARIANCE component, in which case it is the sum of the
model and data variances. i.e the program can be used to compare
two datasets for compatibility. If the data have no error
component then either Poisson or unit errors may be The number of
degrees of freedom is taken to be the number of data points (i.e.
model is assumed to be independent of data). Bad quality data
points are excluded from the statistic, but the model array is
assumed to be 100% good.
Next: FREQUENCY Up: COMPARE Prev: Parameters