Stata 15 help for connect_options

[G-3] connect_options -- Options for connecting data points with lines


connect_options Description ------------------------------------------------------------------------- connect(connectstyle) how to connect points sort[(varlist)] how to sort before connecting cmissing({y|n} ...) missing values are ignored

lpattern(linepatternstyle) line pattern (solid, dashed, etc.) lwidth(linewidthstyle) thickness of line lcolor(colorstyle) color and opacity of line lalign(linealignmentstyle) line alignment (inside, outside, center) lstyle(linestyle) overall style of line

pstyle(pstyle) overall plot style, including linestyle

recast(newplottype) advanced; treat plot as newplottype ------------------------------------------------------------------------- All options are rightmost; see repeated options. If both sort and sort(varlist) are specified, sort is ignored and sort(varlist) is honored.


The connect_options specify how points on a graph are to be connected.

In certain contexts (for example, scatter; see [G-2] graph twoway scatter), the lstyle(), lpattern(), lwidth(), lcolor(), and lalign() options may be specified with a list of elements, with the first element applying to the first variable, the second element to the second variable, and so on. For information about specifying lists, see [G-4] stylelists.


connect(connectstyle) specifies whether points are to be connected and, if so, how the line connecting them is to be shaped; see [G-4] connectstyle. The line between each pair of points can connect them directly or in stairstep fashion.

sort and sort(varlist) specify how the data be sorted before the points are connected.

sort specifies that the data should be sorted by the x variable.

sort(varlist) specifies that the data be sorted by the specified variables.

sort is the option usually specified. Unless you are after a special effect or your data are already sorted, do not forget to specify this option. If you are after a special effect, and if the data are not already sorted, you can specify sort(varlist) to specify exactly how the data should be sorted.

Specifying sort or sort(varlist) when it is not necessary will slow graph down a little. It is usually necessary to specify sort if you specify the twoway option by(), and especially if you include the suboption total.

Options sort and sort(varlist) may not be repeated within the same plot.

cmissing({y|n} ...) specifies whether missing values are to be ignored. The default is cmissing(y ...), meaning that they are ignored. Consider the following data:

+------------+ | rval x | |------------| 1. | .923 1 | 2. | 3.046 2 | 3. | 5.169 3 | 4. | . . | 5. | 9.415 5 | |------------| 6. | 11.538 6 | +------------+

Say that you graph these data by using "line rval x" or equivalently "scatter rval x, c(l)". Do you want a break in the line between 3 and 5? If so, you code

. line rval x, cmissing(n)

or equivalently

. scatter rval x, c(l) cmissing(n)

If you omit the option (or code cmissing(y)), the data are treated as if they contained

+------------+ | rval x | |------------| 1. | .923 1 | 2. | 3.046 2 | 3. | 5.169 3 | 4. | 9.415 5 | 5. | 11.538 6 | +------------+

meaning that a line will be drawn between (3, 5.169) and (5, 9.415).

If you are plotting more than one variable, you may specify a sequence of y/n answers.

lpattern(linepatternstyle), lwidth(linewidthstyle), lcolor(colorstyle), lalign(linealignmentstyle) and lstyle(linestyle) determine the look of the line used to connect the points; see lines. Note the lpattern() option, which allows you to specify whether the line is solid, dashed, etc.; see [G-4] linepatternstyle for a list of line-pattern choices.

pstyle(pstyle) specifies the overall style of the plot, including not only the linestyle, but also all other settings for the look of the plot. Only the linestyle affects the look of line plots. See [G-4] pstyle for a list of available plot styles.

recast(newplottype) is an advanced option allowing the plot to be recast from one type to another, for example, from a line plot to a scatterplot; see [G-3] advanced_options. Most, but not all, plots allow recast().


An important option among all the above is connect(), which determines whether and how the points are connected. The points need not be connected at all (connect(i)), which is scatter's default. Or the points might be connected by straight lines (connect(l)), which is line's default (and is available in scatter). connect(i) and connect(l) are commonly specified, but there are other possibilities such as connect(J), which connects in stairstep fashion and is appropriate for empirical distributions. See [G-4] connectstyle for a full list of your choices.

Equally as important as connect() is sort. If you do not specify this, the points will be connected in the order in which they are encountered. That can be useful when you are creating special effects, but, in general, you want the points sorted into ascending order of their x variable. That is what sort does.

The remaining connect options specify how the line is to look: Is it solid or dashed? Is it red or green? How thick is it? Option lpattern() can be of great importance, especially when printing to a monochrome printer. For a general discussion of lines (which occur in many contexts other than connecting points), see lines.

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index