The mean value, standard deviation, number of values and
the median for the sample [[2 8 7]] are calculated.
By using a sample of values, the "best" fit for the normal dis-
tribution is calculated, i.e. estimates for the mean ^ and
standard deviation a is calculated. Hypothesis testing for as-
sumed distribution is done (kjisquare goodness of Fit).
The sample has to be given as a class statistic with given fre-
quences (se description of samples).
In order to calculate estimates for the the distribution para-
meters a discrete statistic has to be stored as SDAT (see se-
parate menu option).
A "best" fit based on the least squares is calculated. The sam-
ple has to be stored as a class statistic (KSDAT) in advance.
When KsD AT is stored by using a separate menu option, the-
re is no more input data necessary. Estimates for | x and a will