Histogram

Plots histograms (frequency distributions) for one or more columns. You can also use a group column. The number of bins is by default set to an "optimal" number (the zero-stage rule of Wand 1997). The number of bins can be changed by the user.

Optional 95% confidence intervals for the bin counts are computed with the Clopper-Pearson method (see the Single proportion module for computational details).

The "Fit normal" option draws a graph with a fitted normal distribution (Parametric estimation, not Least Squares).

Kernel Density Estimation is a smooth estimator of the histogram. PAST uses a Gaussian kernel with range according to a rule given by Silverman (1986).

Missing values are deleted.

References

Silverman, B.W. 1986. Density estimation for statistics and data analysis. Chapman & Hall.
Wand, M.P. 1997. Data-based choice of histogram bin width. American Statistician 51:59-64.