2D Histograms in MATLAB®

How to make 2D Histogram plots in MATLAB® with Plotly.


Scatter Histogram Chart with Tabular Data

Create a scatter plot with marginal histograms from a table of data for medical patients.

Load the patients data set and create a table from a subset of the variables loaded into the workspace. Then, create a scatter histogram chart comparing the Height values to the Weight values.

load patients
tbl = table(LastName,Age,Gender,Height,Weight);
s = scatterhistogram(tbl,'Height','Weight');

fig2plotly(gcf);

Specify Table Variable for Grouping Data

Using the patients data set, create a scatter plot with marginal histograms and specify the table variable to use for grouping the data.

Load the patients data set and create a scatter histogram chart from the data. Compare the patients' Systolic and Diastolic values. Group the data according to the patients' smoker status by setting the 'GroupVariable' name-value pair argument to 'Smoker'.

load patients
tbl = table(LastName,Diastolic,Systolic,Smoker);
s = scatterhistogram(tbl,'Diastolic','Systolic','GroupVariable','Smoker');

fig2plotly(gcf);

Visualize Categorical and Numeric Data

Use a scatter plot with marginal histograms to visualize categorical and numeric medical data.

Load the patients data set, and convert the Smoker data to a categorical array. Then, create a scatter histogram chart that compares patients' Age values to their smoker status. The resulting scatter plot contains overlapping data points. However, the y-axis marginal histogram indicates that there are far more nonsmokers than smokers in the data set.

load patients
Smoker = categorical(Smoker);
s = scatterhistogram(Age,Smoker);
xlabel('Age')
ylabel('Smoker')

fig2plotly(gcf);

Specify Group Data and Customize Chart Properties

Create a scatter plot with marginal histograms using arrays of shoe data. Group the data according to shoe color, and customize properties of the scatter histogram chart.

Create arrays of data. Then, create a scatter histogram chart to visualize the data. Use custom labels along the x-axis and y-axis to specify the variable names of the first two input arguments. You can specify the title, axis labels, and legend title by setting properties of the ScatterHistogramChart object.

xvalues = [7 6 5 6.5 9 7.5 8.5 7.5 10 8];
yvalues = categorical({'onsale','regular','onsale','onsale', ...
    'regular','regular','onsale','onsale','regular','regular'});
grpvalues = {'Red','Black','Blue','Red','Black','Blue','Red', ...
    'Red','Blue','Black'};
s = scatterhistogram(xvalues,yvalues,'GroupData',grpvalues);

s.Title = 'Shoe Sales';
s.XLabel = 'Shoe Size';
s.YLabel = 'Price';
s.LegendTitle = 'Shoe Color';

fig2plotly(gcf);

Change the colors in the scatter histogram chart to match the group labels. Change the histogram bin widths to be the same for all groups.

xvalues = [7 6 5 6.5 9 7.5 8.5 7.5 10 8];
yvalues = categorical({'onsale','regular','onsale','onsale', ...
    'regular','regular','onsale','onsale','regular','regular'});
grpvalues = {'Red','Black','Blue','Red','Black','Blue','Red', ...
    'Red','Blue','Black'};
s = scatterhistogram(xvalues,yvalues,'GroupData',grpvalues);

s.Title = 'Shoe Sales';
s.XLabel = 'Shoe Size';
s.YLabel = 'Price';
s.LegendTitle = 'Shoe Color';

s.Color = {'Red','Black','Blue'};
s.BinWidths = 1;

fig2plotly(gcf);

Specify Scatter Histogram Chart Appearance

Create a scatter plot with marginal histograms. Specify the number of bins and line widths of the histograms, the location of the scatter plot, and the legend visibility.

Load the patients data set and create a scatter histogram chart from the data. Compare the patients' Diastolic and Systolic values, and group the data according to the patients' SelfAssessedHealthStatus values. Adjust the histograms by specifying the NumBins and LineWidth options. Place the scatter plot in the 'NorthEast' location of the figure by using the ScatterPlotLocation option. Ensure the legend is visible by specifying the LegendVisible option as 'on'.

load patients
tbl = table(LastName,Diastolic,Systolic,SelfAssessedHealthStatus);
s = scatterhistogram(tbl,'Diastolic','Systolic','GroupVariable','SelfAssessedHealthStatus', ...
    'NumBins',4,'LineWidth',1.5,'ScatterPlotLocation','NorthEast','LegendVisible','on');

fig2plotly(gcf);

Group Data Using Two Variables

Create a scatter plot with marginal histograms. Group the data by using a combination of two different variables.

Load the patients data set. Combine the Smoker and Gender data to create a new variable. Create a scatter histogram chart that compares the Diastolic and Systolic values of the patients. Use the new variable SmokerGender to group the data in the scatter histogram chart.

load patients
[idx,genderStatus,smokerStatus] = findgroups(string(Gender),string(Smoker));
SmokerGender = strcat(genderStatus(idx),"-",smokerStatus(idx));
s = scatterhistogram(Diastolic,Systolic,'GroupData',SmokerGender,'LegendVisible','on');
xlabel('Diastolic')
ylabel('Systolic')

fig2plotly(gcf);

Specify Kernel Density Histograms

Create a scatter plot with kernel density marginal histograms. This example requires a Statistics and Machine Learning Toolbox™ license.

Load the carsmall data set and create a scatter histogram chart from the data. Compare the Horsepower and MPG values. Use the number of cylinders to group the data by setting the GroupVariable option to Cylinders. Specify kernel density histograms by setting the HistogramDisplayStyle option to 'smooth'. Specify a solid line for all the histograms by setting the LineStyle option to '-'.

load carsmall
tbl = table(Horsepower,MPG,Cylinders);
s = scatterhistogram(tbl,'Horsepower','MPG', ...
    'GroupVariable','Cylinders','HistogramDisplayStyle','smooth', ...
    'LineStyle','-');

fig2plotly(gcf);

2D Histogram of a Bivariate Normal Distribution

x = randn(500,1);
y = randn(500,1)+1;

data = {...
  struct(...
    'x', x, ...
    'y', y, ...
    'type', 'histogram2d')...
};

plotly(data);

2D Histogram Binning and Styling Options

x = randn(500,1);
y = randn(500,1)+1;

data = {...
  struct(...
    'x', x, ...
    'y', y, ...
    'histnorm', 'probability', ...
    'autobinx', false, ...
    'xbins', struct(...
      'start', -3, ...
      'end', 3, ...
      'size', 0.1), ...
    'autobiny', false, ...
    'ybins', struct(...
      'start', -2.5, ...
      'end', 4, ...
      'size', 0.1), ...
    'colorscale', { { {0, 'rgb(12,51,131)'},{0.25, 'rgb(10,136,186)'},{0.5, 'rgb(242,211,56)'},{0.75, 'rgb(242,143,56)'},{1, 'rgb(217,30,30)'} } }, ...
    'type', 'histogram2d')...
};

plotly(data);

2D Histogram Overlaid with a Scatter Chart

x0 = randn(100,1)./5. + 0.5;
y0 = randn(100,1)./5. + 0.5;
x1 = rand(50,1);
y1 = rand(50,1) + 1.0;

x = [x0; x1];
y = [y0; y1];

trace1 = struct(...
        'x', x0, ...
        'y', y0, ...
        'mode', 'markers', ...
        'marker', struct(...
          'symbol', 'circle', ...
          'opacity', 0.7), ...
        'type', 'scatter');

trace2 = struct(...
        'x', x1, ...
        'y', y1, ...
        'mode', 'markers', ...
        'marker', struct(...
          'symbol', 'square', ...
          'opacity', 0.7), ...
        'type', 'scatter');

trace3 = struct(...
        'x', x, ...
        'y', y, ...
        'type', 'histogram2d');

data = {trace1, trace2, trace3};

layout = struct(...
          'legend', struct(...
        'y', 1.0, ...
        'yanchor', 'bottom', ...
        'x', 1.0, ...
        'xanchor', 'right'), ...
          'xaxis', struct('zeroline', false), ...
          'yaxis', struct('zeroline', false) ...
          );

plotly(data, struct('layout', layout));