rootgram Hanging rootograms for discrete distributions rootgram

Visualizing Categorical Data: rootgram

$Version: 1.0 (9 Jan 2001)
Michael Friendly
York University

The rootgram macro ( [download] get rootgram.sas)

Hanging rootograms for discrete distributions

The ROOTGRAM macro produces histograms, rootograms, and hanging rootograms for the distribution of a discrete variable compared with expected frequencies according to a theoretical distribution.

Usage

The VAR= and OBS= variables must be specified. The expected frequencies may be obtained with the GOODFIT macro.

The ROOTGRAM macro is called with keyword parameters. The arguments may be listed within parentheses in any order, separated by commas. For example:

  %include catdata(madison);
  %goodfit(data=madison, var=count, freq=blocks, dist=poisson);
  %rootgram(data=fit, var=count, obs=blocks);

Parameters

DATA=
Specifies the name of the input data set [Default: DATA=_LAST_]
VAR=
Specifies the name of the analysis variable, used as the abscissa in the plot.
OBS=
Specifies the bserved frequency variable
EXP=
Expected/fitted frequency [Default: EXP=EXP]
FUNC=
The name of the function applied to ordinate. Use FUNC=NONE to give a plot on the scale of observed frequency. [Default: FUNC=SQRT]
BWIDTH=
Bar width [Default: BWIDTH=.5]
BCOLOR=
Bar color [Default: BCOLOR=GRAYB0]
BTYPE=
Bar type: One of HANG, DEV, or NEEDLE. [Default: BTYPE=HANG]
ANNO=
The name of an input annotate dataset
NAME=
Name of the graphics catalog entry

Example

%include vcd(rootgram);        *-- or include in an autocall library;
%include vcd(goodfit);         *-- or include in an autocall library;

data saxony;
	do males = 0 to 12;
		input families @;
		output;
		end;
	label males='Number of males' families='Number of families';
cards;
3  24  104  286  670  1033  1343 1112  829  478  181  45  7
;
%goodfit(data=saxony, var=males, freq=families, dist=binomial);
%rootgram(data=fit, var=males, obs=families, exp=exp);

See also

distplot Plots for discrete distributions
goodfit Goodness of fit tests for discrete distributions
ordplot Diagnose form of discrete frequency distribution
poisplot Poissonness plot for discrete distributions

[Previous] [Next] [Up] [Contents] [Search]
Previous: robust Next: sieve Up: Appendix A Top: index.html