gplotmatrix

Matrix of scatter plots by group

Syntax

gplotmatrix(X,[],group)

gplotmatrix(X,Y,group)

gplotmatrix(X,Y,group,clr,sym,siz)

gplotmatrix(X,Y,group,clr,sym,siz,doleg)

gplotmatrix(X,[],group,clr,sym,siz,doleg,dispopt)

gplotmatrix(X,[],group,clr,sym,siz,doleg,dispopt,xnam)

gplotmatrix(X,Y,group,clr,sym,siz,doleg,[],xnam,ynam)

gplotmatrix(parent,___)

[h,ax,bigax] = gplotmatrix(___)

Description

gplotmatrix(X,[],group) creates a matrix of scatter plots and histograms of the data in X, grouped by the grouping variable in group. Each off-diagonal plot in the resulting figure is a scatter plot of a column of X against another column of X. The software also plots the outlines of the grouped histograms in the diagonal plots of the plot matrix. X and group must have the same number of rows.

gplotmatrix(X,Y,group) creates a matrix of scatter plots. Each plot in the resulting figure is a scatter plot of a column of X against a column of Y. For example, if X has p columns and Y has q columns, then the figure contains a q-by-p matrix of scatter plots. All plots are grouped by the grouping variable group. The input arguments X, Y, and group must all have the same number of rows.

gplotmatrix(X,Y,group,clr,sym,siz) specifies the marker color clr, symbol sym, and size siz for each group.

gplotmatrix(X,Y,group,clr,sym,siz,doleg) controls whether a legend is displayed in the figure. gplotmatrix creates a legend by default.

gplotmatrix(X,[],group,clr,sym,siz,doleg,dispopt) controls the display options for the diagonal plots in the plot matrix of X.

gplotmatrix(X,[],group,clr,sym,siz,doleg,dispopt,xnam) labels the x-axes and y-axes of the scatter plots using the column names specified in xnam. The input argument xnam must contain one name for each column of X. Set dispopt to 'variable' to display the variable names along the diagonal of the scatter plot matrix.

gplotmatrix(X,Y,group,clr,sym,siz,doleg,[],xnam,ynam) labels the x-axes and y-axes of the scatter plots using the column names specified in xnam and ynam. The input arguments xnam and ynam must contain one name for each column of X and Y, respectively.

gplotmatrix(parent,___) creates the scatter plot matrix in the figure or panel specified by parent. Specify parent as the first input argument followed by any of the input argument combinations in the previous syntaxes.

[h,ax,bigax] = gplotmatrix(___) returns graphics handles to the individual plots and the entire scatter plot matrix.

You can pass in [] for clr, sym, siz, doleg, and dispopt to use their default values.

Examples

Scatter Plots with Grouped Data

Create a matrix of scatter plots for each combination of variables in a data set. Group the data according to a separate variable.

Load the fisheriris data set, which contains flower data. The four columns of meas are the sepal length, sepal width, petal length, and petal width of the flowers. species contains the flower species names: setosa, versicolor, and virginica. Visually compare the flower measurements across flower species.

load fisheriris
gplotmatrix(meas,[],species)

In the matrix of scatter plots, the x-axis of the leftmost column of scatter plots corresponds to sepal length, the first column in meas. Similarly, the y-axis of the bottom row of scatter plots corresponds to petal width, the last column in meas. Therefore, the scatter plot in the bottom left of the matrix compares sepal length values (along the x-axis) to petal width values (along the y-axis). The color of each point depends on the species of the flower.

The diagonal plots are histograms rather than scatter plots. For example, the plot in the top left of the matrix shows the distribution of sepal length values for each species of flower.

Create Scatter Plot Matrix with Subset of Variables

Create scatter plots comparing a subset of the variables in a data set to another subset of variables. Group the data according to a separate variable.

Load the discrim data set.

load discrim

The ratings array contains rating values of 329 US cities for the nine categories listed in the categories array. The group array contains a city size code that is equal to 2 for the 26 largest cities, and 1 otherwise.

Create a matrix of scatter plots to compare the first two categories, climate and housing, with the fourth and seventh categories, crime and arts. Specify group as the grouping variable to visually distinguish the data for large and small cities.

X = ratings(:,1:2);
Y = ratings(:,[4 7]);
gplotmatrix(X,Y,group)

The matrix of scatter plots shows the specified comparisons, with each city size group represented by a different color.

Adjust the appearance of the plots by specifying marker colors and symbols, and labeling the axes with the rating categories.

xnames = categories(1:2,:);
ynames = categories([4 7],:);
gplotmatrix(X,Y,group,'br','.o',[],'on',[],xnames,ynames)

Scatter Plot Matrix with Multiple Grouping Variables

Create a matrix of scatter plots comparing data variables by using two grouping variables.

Load the patients data set. Compare patient diastolic and systolic blood pressure values. Group the patients according to their gender and smoker status. Convert Smoker to a categorical variable to have more descriptive labels in the legend. Display grouped histograms along the diagonal of the plot matrix by using the 'grpbars' display option, and label the axes.

load patients
X = [Diastolic Systolic];
labeledSmoker = categorical(Smoker,[true false],{'Smoker','Nonsmoker'});
group = {Gender,labeledSmoker};
color = lines(4)

color = 4×3

         0    0.4470    0.7410
    0.8500    0.3250    0.0980
    0.9290    0.6940    0.1250
    0.4940    0.1840    0.5560

xnames = {'Diastolic','Systolic'};
gplotmatrix(X,[],group,color,[],[],[],'grpbars',xnames)

For example, the scatter plot in the bottom left of the matrix shows that smokers (blue and yellow markers) tend to have higher diastolic and systolic blood pressure values, regardless of gender.

Modify Scatter Plot Matrix Appearance

Create a matrix of scatter plots that display grouped data. Modify the appearance of one of the scatter plots.

Load the carsmall data set. Create a scatter plot matrix using different car measurements. Group the cars by the number of cylinders. Specify the group colors, and display the car variable names along the diagonal of the plot matrix. Add a title to the plot matrix.

load carsmall
X = [Acceleration Displacement Horsepower MPG Weight];
color = lines(3)

color = 3×3

         0    0.4470    0.7410
    0.8500    0.3250    0.0980
    0.9290    0.6940    0.1250

xnames = {'Acceleration','Displacement','Horsepower','MPG','Weight'};
[h,ax] = gplotmatrix(X,[],Cylinders,color,[],[],[],'variable',xnames);
title('Car Data')

Change the appearance of the scatter plot in the bottom left of the matrix by using h and ax. First, change the colors of the data points in the scatter plot. Then, add grid lines to the scatter plot.

bottomleftPlot = h(5,1,:);
bottomleftPlot(1).Color = 'blue';
bottomleftPlot(2).Color = 'red';
bottomleftPlot(3).Color = 'yellow';

bottomleftAxes = ax(5,1);
bottomleftAxes.XGrid = 'on';
bottomleftAxes.YGrid = 'on';

Input Arguments

X — Input data

numeric matrix | datetime array | duration array

Input data, specified as an n-by-p numeric matrix, datetime array, or duration array. gplotmatrix creates a matrix of plots using the columns of X. If you do not specify an additional input matrix Y, then gplotmatrix creates a p-by-p matrix of plots. The off-diagonal plots are scatter plots, and the diagonal plots depend on the value of dispopt. In each scatter plot, gplotmatrix plots one column of X against another column of X. The points in the scatter plots are grouped according to group.

If you specify Y, then gplotmatrix creates a q-by-p matrix of scatter plots using the p columns of X and the q columns of Y.

Data Types: single | double | datetime | duration

Y — Input data

numeric matrix | datetime array | duration array

Input data, specified as an n-by-q numeric matrix, datetime array, or duration array. gplotmatrix creates a q-by-p matrix of scatter plots using the p columns of X and the q columns of Y. For each column of the plot matrix, the x-axis values of the scatter plots are the same as the values in the corresponding column of X. Similarly, for each row of the plot matrix, the y-axis values of the scatter plots are the same as the values in the corresponding column of Y. The points in the scatter plots are grouped according to group.

X and Y must have the same number of rows.

Data Types: single | double | datetime | duration

group — Grouping variable

categorical vector | numeric vector | logical vector | character array | string array | cell array

Grouping variable, specified as a categorical vector, numeric vector, logical vector, character array, string array, or cell array of character vectors. Alternatively, group can be a cell array containing several grouping variables (such as {g1 g2 g3}), in which case observations are in the same group if they have common values of all grouping variables. In any case, group must have the same number of rows as X. Points in the same group appear on the graph with the same marker color, symbol, and size.

Example: categorical({'blue','red','yellow','blue','yellow','red','red','yellow','blue','red'})

Example: {Smoker,Gender} where Smoker and Gender are grouping variables

Data Types: categorical | single | double | logical | char | string | cell

clr — Marker colors

character vector | string scalar | string array | cell array of character vectors | matrix of RGB values

Marker colors, specified as one of the following:

• Character vector or string scalar of color short names.

• String array or cell array of character vectors designating color names or short names.

• Three-column matrix of RGB values in the range [0,1]. The three columns represent the R (red) value, G (green) value, and B (blue) value.

You can choose among these predefined colors and their equivalent RGB triplets.

Color Name	Short Name	RGB Triplet	Appearance
'red'	'r'	[1 0 0]
'green'	'g'	[0 1 0]
'blue'	'b'	[0 0 1]
'cyan'	'c'	[0 1 1]
'magenta'	'm'	[1 0 1]
'yellow'	'y'	[1 1 0]
'black'	'k'	[0 0 0]
'white'	'w'	[1 1 1]

When the total number of groups exceeds the number of specified colors, gplotmatrix cycles through the specified colors.

Example: {'blue','black','green'}

Example: [0 0 1; 0 0.5 0.5; 0.5 0.5 0.5]

Data Types: char | string | cell | single | double

sym — Marker symbols

'.' (default) | character vector | string scalar

Marker symbols, specified as a character vector or string scalar.

You can choose among these marker options.

Value	Description
'o'	Circle
'+'	Plus sign
'*'	Asterisk
'.'	Point
'x'	Cross
's'	Square
'd'	Diamond
'^'	Upward-pointing triangle
'v'	Downward-pointing triangle
'>'	Right-pointing triangle
'<'	Left-pointing triangle
'p'	Five-pointed star (pentagram)
'h'	Six-pointed star (hexagram)
'none'	No markers

By default, gplotmatrix assigns '.' as the marker symbol for each group. When the total number of groups exceeds the number of specified symbols, gplotmatrix cycles through the specified symbols.

Example: 'x'

Example: 'xo+'

Data Types: char | string

siz — Marker sizes

positive numeric vector

Marker sizes, specified as a positive numeric vector. The default value is determined by the number of observations. When the total number of groups exceeds the number of specified sizes, gplotmatrix cycles through the specified sizes.

Example: [6 12]

Data Types: single | double

doleg — Option to include legend

'on' (default) | 'off'

Option to include a legend, specified as either 'on' or 'off'. By default, the legend is displayed in the figure.

dispopt — Display options for diagonal plots

'stairs' (default) | 'hist' | 'grpbars' | 'none' | 'variable'

Display options for the diagonal plots in the plot matrix, specified as 'stairs', 'hist','grpbars', 'none', or 'variable'. This table describes the different display options.

Value	Description
'stairs'	Plot the outlines of grouped histograms.
'hist'	Plot histograms.
'grpbars'	Plot grouped histograms.
'none'	Display blank plots.
'variable'	Display variable names. To use this display option, you must specify xnam.

The default is 'stairs' when group contains more than one group. Otherwise, gplotmatrix displays a single histogram in each diagonal plot.

To generate the histograms, gplotmatrix uses the 'pdf' type of normalization for numeric data and the 'count' type of normalization for datetime and duration data. (See the 'Normalization' name-value pair argument of histogram.) Note that the y-axis tick mark labels do not apply to the histograms. Use data tips to see the correct histogram values.

xnam — X column names

character array | string array | cell array of character vectors

X column names, specified as a character array, string array, or cell array of character vectors. xnam must contain one name for each column of X.

Example: {'Sepal Length','Sepal Width','Petal Length','Petal Width'}

Data Types: char | string | cell

ynam — Y column names

character array | string array | cell array of character vectors

Y column names, specified as a character array, string array, or cell array of character vectors. ynam must contain one name for each column of Y.

Example: {'Diastolic','Systolic'}

Data Types: char | string | cell

parent — Parent container

Figure object | Panel object

Parent container, specified as a Figure or Panel object.

Output Arguments

h — Line handles to individual plots

array of Line and Histogram objects

Line handles to individual plots, returned as a one of these arrays:

• p-by-p-by-k array of Line and Histogram objects if you do not specify Y

• q-by-p-by-k array of Line objects if you specify both X and Y

p is the number of columns in X, q is the number of columns in Y, and k is the number of unique groups in group.

Each scatter plot has k corresponding Line objects in h, and each histogram has k corresponding Histogram objects in h.

ax — Axes handles to individual plots

matrix of Axes objects

Axes handles to individual plots, returned as a matrix of Axes objects. If dispopt is 'hist', 'stairs', or 'grpbars', then ax contains one extra row of handles to invisible axes where the histograms are plotted.

bigax — Axes handle to entire plot matrix

Axes object

Axes handle to the entire plot matrix, returned as an Axes object. bigax points to the current axes, so a subsequent title, xlabel, or ylabel command produces labels that are centered with respect to the entire plot matrix.