HISTOGRAM Procedure

@HISTOGRAM computes and displays a density histogram for a series.

@Histogram( options ) series start end

Parameters

series	series to analyze
start, end	range to use. By default, the defined range of series

Options

DISTRIB=[NONE]/NORMAL

For DISTRIB=NORMAL, shows an overlay of the normal density.

STATS/[NOSTATS]

With STATS, displays series statistics in the margin

HLABEL="horizontal axis label" [not used]

VLABEL="vertical axis label" [not used]

HEADER="header label" [not used]

FOOTER="footer label" [not used]

COUNTS/[NOCOUNTS]

Computes and displays counts in each bin rather than density

MAXGRID=maximum number of bins [25]

VMAX=maximum value for graph

XGRID=series for grid boundaries [not used]

Example

* Pindyck & Rubinfeld, Econometric Models and Economic Forecasts, 4th edition

* Example 3.5 from page 78

open data ex35.xls

data(format=xls,org=columns) 1 50 obs enb env

linreg enb

# constant env

stats(fractiles) %resids

@histogram(footer="Figure 3.6 Histogram of Residuals",$

hlabel="Residual per thousand enrollment",$

vlabel="Number of States",maxgrid=16,counts) %resids

Sample Graph

This is from the example above. With just 50 data points, 16 grid points is really too many which shows in the graph.