Spider Plot with Standard Deviation as shaded region

I want to plot a spider plot where each spoke represents the average value with solid line. In addition, I want to show standard deviation as shaded region around the average plot (Average+SD and Average-SD). I have attached a figure to show the desired outcome.

Figure available at: https://www.mdpi.com/2078-2489/15/6/364

Thank you in advance.

[Mathieu NOE]

you can probably add the shaded area in this existing code :

spider_plot - File Exchange - MATLAB Central

[Ayesha Tooba Khan]

Yes, I am using this code only. But, if I am keeping 'fill' as 'on' it is filling the etire region starting from the origin. I only want to have SD region as shaded one.

Thank you for the response.

Answer 1

[Mathieu NOE]

hello again

I just created a quick and dirty slightly modified function (sorry for messing it ! ) , so I could suggest this - if this is what you want

for the demo I assumed that each D vector has a correspond std vector called stdD which here is 10% of its nominal D value (of course you can put your own data in here)

custom function is attached (my apologizes to it's creator !) (NB : I simply deactivated certain checks which would otherwise throw error messages)

my result :

% Initialize data points
D1 = [5 3 9 1 2];
D2 = [5 8 7 2 9];
D3 = [8 2 1 4 6];
P = [D1; D2; D3];

% Multiple shaded regions
stdD1 = 0.1*D1;
stdD2 = 0.1*D2;
stdD3 = 0.1*D3;
axes_shaded_limits = {...
    [D1-stdD1;D1+stdD1],... % [min axes limits; max axes limits]
    [D2-stdD2;D2+stdD2],...
    [D3-stdD3;D3+stdD3]};

% Spider plot
spider_plot_mod1(P,...
    'AxesShaded', 'on',...
    'AxesShadedLimits', axes_shaded_limits,...
    'AxesShadedColor', {'b', 'r','y'},...
    'AxesShadedTransparency', 0.1);

[Ayesha Tooba Khan]

Hello,

While implementing this code on my data, I am receiving the following warnings and then a figure is getting plotted which doesn't contain data points:

Warning: Output contains an empty polyshape due to duplicate vertices, intersections, or other

inconsistencies.

Warning: Polyshape has duplicate vertices, intersections, or other inconsistencies that may produce

inaccurate or unexpected results. Input data has been modified to create a well-defined polyshape.

I am not able to figure out the reason for it.

I will be grateful if you can kindly look into the issue. I am enclosing the code, figure and the data.

Thank you so much.

[Mathieu NOE]

ok I understand - as I learn how to use the function spider_plot

To avoid this situation you have to use the argument AxesLimits

here for example the same code I used above in my first answer, but with now the axis limits argument

NB it's a different argument , not to confuse with AxesShadedLimits

see the plot has now all axes in range 0 to 10 as specified by argument AxesLimits

% Initialize data points
D1 = [5 3 9 1 2];
D2 = [5 8 7 2 9];
D3 = [8 2 1 4 6];
P = [D1; D2; D3];

% Multiple shaded regions
stdD1 = 0.1*D1;
stdD2 = 0.1*D2;
stdD3 = 0.1*D3;

% shaded area limits
axes_shaded_limits = {...
    [D1-stdD1;D1+stdD1],... % [min axes limits; max axes limits]
    [D2-stdD2;D2+stdD2],...
    [D3-stdD3;D3+stdD3]};

% axes limits
AxesLimits_min = zeros(1,size(P,2));
AxesLimits_max = ceil(max(P(:))+1)*ones(1,size(P,2));
AxesLimits = [AxesLimits_min; AxesLimits_max];

% Spider plot
spider_plot_mod1(P,...
    'AxesShaded', 'on',...
    'AxesLimits',AxesLimits,...
    'AxesShadedLimits', axes_shaded_limits,...
    'AxesShadedColor', {'b', 'r','y'},...
    'AxesShadedTransparency', 0.3);

[Mathieu NOE]

now if I transpose with your data / code , this would be :

load('NormalizedVoluntaryAverageFeaturesWithSD.mat');
Average=AverageVoluntaryAverageNMR(:,:,:,1);
SD = SDVoluntaryAverageNMR(:,:,:,1);
UpperBound = Average+SD;
LowererBound = Average-SD;

% shaded area limits
axes_shaded_limits = {[LowererBound;UpperBound]};

% axes limits
AxesLimits_min = floor(min(Average(:)))*ones(1,size(Average,2));
AxesLimits_max = ceil(max(Average(:)))*ones(1,size(Average,2));
AxesLimits = [AxesLimits_min; AxesLimits_max];

% Spider plot
spider_plot_mod1(Average,...
    'AxesShaded', 'on',...
    'AxesLimits',AxesLimits,...
    'AxesShadedLimits', axes_shaded_limits,...
    'AxesShadedColor', {'b'},...
    'AxesShadedTransparency', 0.3);

[Mathieu NOE]

if you want to "zoom" the axes range , you can change the definition of AxesLimits

like here :

load('NormalizedVoluntaryAverageFeaturesWithSD.mat');
Average=AverageVoluntaryAverageNMR(:,:,:,1);
SD = SDVoluntaryAverageNMR(:,:,:,1);
UpperBound = Average+SD;
LowererBound = Average-SD;

% shaded area limits
axes_shaded_limits = {[LowererBound;UpperBound]};

% axes limits
% AxesLimits_min = floor(min(Average(:)))*ones(1,size(Average,2));
% AxesLimits_max = ceil(max(Average(:)))*ones(1,size(Average,2));
AxesLimits_min = (min(Average(:))-0.1)*ones(1,size(Average,2));
AxesLimits_max = (max(Average(:))+0.1)*ones(1,size(Average,2));
AxesLimits = [AxesLimits_min; AxesLimits_max];

% Spider plot
spider_plot_mod1(Average,...
    'AxesShaded', 'on',...
    'AxesLimits',AxesLimits,...
    'AxesShadedLimits', axes_shaded_limits,...
    'AxesShadedColor', {'b'},...
    'AxesShadedTransparency', 0.3);

[Ayesha Tooba Khan]

Thank you so much. It is perfectly fine now.

[Mathieu NOE]

hello

there is a small change in the code that is not evident to see , but it makes a difference in term of size() and that is the reason why it didn't work in first place

the correct way to generate the axes_shaded_limits :

axes_shaded_limits = {[LowerBoundUp;LowerBoundDown],...
    [UpperBoundUp;UpperBoundDown]};    % shaded area limits

and this has size : axes_shaded_limits = 1×2 cell array, {2×8 double} {2×8 double}

the way you coded is not equivalent :

    UpperBound = [UpperBoundUp;UpperBoundDown];
    LowerBound = [LowerBoundUp;LowerBoundDown];
    axes_shaded_limits = {[LowerBound,...
        UpperBound]};

will generate this output : axes_shaded_limits = 1×1 cell array, {2×16 double}

which is incorrect (in size) and will generate the error msg

Updated code is attached

all the best

[Ayesha Tooba Khan]

Yes, I got stuck there but got that resolved.

just a little change: according to my problem, the axes_shaded_limit would be

axes_shaded_limits = {[LowerBoundUp;UpperBoundUp],...

[LowerBoundDown;UpperBoundDown]};

as the average will have shaded region of lower and upper bounds (for Up and Down condition which corresponds to spider plots on a single tile, in my case).

Just wanted to clarify, if anyone wants to use the same in future.

I sincerely appreciate your help. Thank you so much.

[Mathieu NOE]

my pleasure !

Answer 2

[Thorsten]

You can do it like this:

I am not aware of a function, so you have to do it on your own.

Note that this functions uses values maxval = 7 and offset = 4 that are tailored to your data to have an axis running from -4 to 3.

Your friend is pol2cart to convert all the polar stuff to Cartesian coordinates and then use plot, patch or line commands.

function spiderplot(val, sd, col)

if nargin < 1
    val = [-1 -1 1 -1 -1 -1 0];
end   
if nargin < 2
    Nspokes = size(val, 2);
    sd = ones(1, Nspokes) + 0.5*rand(1, Nspokes);
end

if nargin < 3
  col = 'r'; % or [234 170 58]/266; or 'g' or any other color 
end

labels = {'Degree', 'Density', 'Betweeness Centrality', 'Closeness', 'Eigenvector', 'Transivity', 'Entropy'};

offset = 4; % map the smallest value to 0
maxval = 7;

val = val + offset;

Nspokes = size(val, 2);
theta = linspace(0, 2*pi, Nspokes + 1);

%% plot the axes (background, circles and spokes)
% add a grayish background circle
col_bg = [229, 236, 246]/255;
Npoints = 100;
thetaN = linspace(0, 2*pi, Npoints);
[xc, yc] = pol2cart(thetaN, maxval);
patch(xc, yc, col_bg, 'EdgeColor', 'none')
hold on

% add the spokes
for i = 1:Nspokes
    [xs, ys] = pol2cart(theta(i), maxval);
    line([0, xs], [0, ys], 'Color', 'w')
end
% add the circles
for i = 1:maxval
    [xc, yc] = pol2cart(thetaN, i);
    plot(xc, yc, 'w-')
end
% label the axis
for i = 0:maxval
    str = int2str(i - offset);
    % change dash to minus:
    str = strrep(str, '-', char(hex2dec('2212'))); 
    text(i, 0, str, 'FontSize', 14,...
        'HorizontalAlignment', 'center', 'VerticalAlignment', 'top')
end

axis off

%% plot the data
% append first element to plot a closed curve
val = [val, val(:, 1)];
sd = [sd, sd(:, 1)];

% plot data
[x, y] = pol2cart(theta, val); % convert to Cartesian
plot(x, y, '-', 'Color', col)
hold on

% plot sd lines
[x1, y1] = pol2cart(theta, val - sd);
[x2, y2] = pol2cart(theta, val + sd);

plot(x1, y1, '--', 'Color', col)
plot(x2, y2, '--', 'Color', col)

% shade the area between the +/- SD lines
h = patch([x1 fliplr(x2)], [y1 fliplr(y2)], col,...
    'EdgeColor', 'none', 'FaceAlpha', 0.2);

[Ayesha Tooba Khan]

Thank you so much for considering even the minor details of the example figure.