------------------------------------------------------------------------------- help for cvxplot R Allan Reese, University of Hull, UK ------------------------------------------------------------------------------- program cvxplot version 8.0 syntax varlist(min=4) [if] [in] /// [ , Hulls(numlist int min=0 max=2 >0) MDPlot(int 8) /// GROup(varname) SELect(numlist int min=0 >0 sort) /// noREPort MEAns SCATopt(string) SAVing(string)]

Draws convex hulls on scatterplot, using values returned by cvxhull.

ANY OTHER USE IS ENTIRELY AT THE USER'S RISK

cvxplot yvar xvar list of hulls [if exp] [in range] [, hulls(int[, int]) group(varname) select(numlist) noreport means scatopt(string) saving(graph_filename[, replace])]

Description

cvxplot makes a scatterplot and draw convex hulls of a group of points in two-dimensional space. Each hull is defined by two lines joining the bottom-left point to the top-right point. The observations do not need to be sorted before calling the routine, but the group variable must match the one used in setting up the hulls. The coding used for each hull is as set up by cvxplot, the "left" and "right" vectors containing the "y" values for a line.

Typical use may be to call cvxhull to calculate many hulls and save them, then call cvxplot to draw selected hulls for selected groups.

eg cvxhull yv xv, hulls(50) group(gv) mdp(5) prefix(hn) saving(first5) cvxplot yv xv hn1l-hn50r, hulls(50 10) group(gv) saving(by10) cvxplot yv xv hn1l-hn50r, h(50 2) gro(gv) sav(odds) cvxplot yv xv hn2l-hn50r, h(50 2) gro(gv) sav(evens)

Parameters

There must be a varlist with an even number of names. The first two are the y and x; the remainder are pairs of variables for each layer of hull. The routine cvxhull saves the hull variables in a block so that normally you need refer only to first-last.

The if and in qualifiers act as usual to prescribe a subset of observations to be considered. It is left to the user to decide if a selection different from the one in force for cvxhull gives sensible results.

Options

Main options

hulls(numlist) one or two integers that define the maximum number of hulls to be plotted from the varlist. Defaults to 1. If the second integer has a value n, only the first and every n'th hull from the varlist are plotted. Some spurious comments relating to pen styles may be output for hull levels when large numbers of hulls are defined. These appear to be harmless.

group(varname) separate hulls are defined for each set of points identified with a different value of varname. The variable may be numeric or string. A table is generated showing the equivalence of numbers with the original values.

select(numlist) only groups indexed in the list will have hulls drawn, but other groups will be plotted as points. Different plots may therefore be obtained by the commands

cvxplot y x if group==1 and cvxplot y x, group(group) select(1)

means group means of y-x will be added to the graph as solid triangles.

scatopt(string) provides a string of options that will be applied to the scatterplot. Any scatterplot options can be applied, obvious examples being scatopt(ms(i)) to make the points invisible leaving only the hull outlines, scatopt(mlab(var)) to mark each point with a value. {cmd:noreport suppresses the progress messages while the program is running.

saving(graph_filename[,replace]) saves the graph in a .gph file.

Examples

. cvxplot yvar xvar _cvxh1l _cvxh1r Draws a default scatterplot with one convex hull containing all the observations using cvxhull's standard names.

. cvxhull y x, h(5) group(year) prefix(hl) scat(mlab(month)) . cvxplot y x hl1l-hl5r, group(year) scat(ms(i)) h(5) The first call computes 5 hulls but draws just one for each year and the month identifiers will be enclosed by a boundary. The second call shows all the hulls but not the points.

. gen lab = _n if hlpts==1 . cvxplot y x hl1l-hl5r, group(year) prefix(hl) h(2) scat(mlab(lab)) Draws two levels of hull but labels only the points on the outer one.

Author

R Allan Reese, University of Hull, UK r.a.reese@hull.ac.uk

Also see

On-line: help for cvxhull graph scatter Manual: [R] graph, [R] connectlines