.-
help for ^cdf^                        (STB-49: gr37)
.-

Cumulative distribution function
--------------------------------

^cdf^ varname [weight] [^in^ range] [^if^ exp], [^by(^varname^)^] [^norm^al] [^same^sd]


Description
-----------

^cdf^ plots the sample cumulative distribution function. Distributions can
be compared within subgroups defined by a second variable. The best fitting 
normal (Gaussian) model may be superimposed over the sample c.d.f.


Options
-------

^by(^varname^)^ causes a separate c.d.f. to be calculated for each value 
of varname. These are plotted on the same graph for easier comparison.

^norm^al causes a normal probability curve with the same mean and standard 
deviation to be superimposed over the c.d.f.

^same^sd is relevant only when ^by^ and ^normal^ options are used together.
It causes normal curves with different means but the same standard deviations
to be fitted. This demonstrates the fit of the conventional Gaussian location
shift model.
 
^graph_options^ are allowed. Default labelling is supplied when graph_options
are absent, but the x-axis label may be supplied in the ^b2^ graphics 
option and the y-axis may be labelled using the ^l1^ option. If the 
^xlog^ option is used, the ^normal^ option causes log-normal distributions
to be fitted.


Examples
--------

Using the variable ^length^ from the ^auto^ data:

^cdf length, normal^
^cdf length, by(foreign)^
^cdf length, by(foreign) norm b2("Length of auto") saving(mygraph)^


Authors
-------

      David Clayton
      MRC Biostatistical Research Unit, Cambridge
      david.clayton@@mrc-bsu.cam.ac.uk

      Michael Hills (retired)
      mhills@@regress.demon.co.uk


Also see
--------

     STB: STB-49 gr37