Subplots of subplots (maybe using subaxis)

I am using the beast exchange file, which default example,

load('Nile.mat')                  % annual streamflow of the Nile River startin from year 1871
out = beast(Nile,'start', 1871);  % trend and seasonality of the time-series
H = plotbeast(out)

produces the following subplots:

H = 
    1×8 Axes array:
    Axes    Axes    Axes    Axes    Axes    Axes    Axes    Axes

Now, I would like to create subplots of those subplots, produced by beast, possibly using the subaxis function (but I am open to other tools for subplots!). Unfortunately, the following attempt does not work:

load('Nile.mat') 
out = beast(Nile,'start', 1871);
% subplot of subplots?
for i = 1 : 6
    subax(i) = subaxis(4,2,i); 
    H = plotbeast(out)
end

My goal would be to get something like this:

How can I get it ? Do you have any suggestion / idea?

[Sim]

hey guys...any idea? I am bit sinking with this plot :-) :-)

[Kaiguang]

%install the beast program
eval(webread('http://b.link/beast',weboptions('cert','')))

% load a sample time series
load('Nile.mat')

% For this example, make sure not to fit the seasonality bcz it is an
% annual trend-only streamflow time series, without any periodic component
o=beast(Nile,'season','none'); 

% o is a struct variable containing all the results 
o  

% Plot the results directly from o
clf
subplot(3,1,1)
plot(o.time, o.data)  % plot the original data

subplot(3,1,2)         
plot(o.time, o.trend.Y)  % plot the fittted trend

subplot(3,1,3)         
plot(o.time, o.trend.cpOccPr)  % plot the changepoint occurance probability


% if needed, check this pdf https://cran.r-project.org/web/packages/Rbeast/Rbeast.pdf
% starting from Page 7 to see more details about the output fields 


%Alternatively, use the plotbeast to choose the subplots through the 'vars'
%argument

plotbeast(o,'vars',["t","tcp","torder"])
plotbeast(o,'vars',["t","tcp","torder","error"])
plotbeast(o,'vars',["t","error"])


 

[Kaiguang]

Shown above are some examples to plot the results from the beast outputs. If more help is needed, contact me at [e-mail].

[Sim]

thanks a lot @Kaiguang, very kind!!!

Answer 1

[Kaiguang]

Repost my code snippet here to explain the beast output:

%install the beast program
eval(webread('http://b.link/beast',weboptions('cert','')))
% load a sample time series
load('Nile.mat')
% For this example, make sure not to fit the seasonality bcz it is an
% annual trend-only streamflow time series, without any periodic component
o=beast(Nile,'season','none'); 
% o is a struct variable containing all the results 
o  
% Plot the results directly from o


clf
subplot(3,1,1)
plot(o.time, o.data)  % plot the original data
subplot(3,1,2)         
plot(o.time, o.trend.Y)  % plot the fittted trend
subplot(3,1,3)         
plot(o.time, o.trend.cpOccPr)  % plot the changepoint occurance probability
% if needed, check this pdf https://cran.r-project.org/web/packages/Rbeast/Rbeast.pdf
% starting from Page 7 to see more details about the output fields 
%Alternatively, use the plotbeast to choose the subplots through the 'vars'
%argument
plotbeast(o,'vars',["t","tcp","torder"])
plotbeast(o,'vars',["t","tcp","torder","error"])
plotbeast(o,'vars',["t","error"])

[Sim]

thanks a lot @Kaiguang!!

Answer 2

[the cyclist]

I would try the native tiledlayout function to make the subplots. I've never used beast, so I'm not sure if there are any weird incompatibilities.

[Sim]

thanks @the cyclist :) But, unfortunately, it does not work...

load('Nile.mat')                  
out = beast(Nile,'start', 1871);  

tiledlayout(2,1);

% Tile 1
nexttile
plotbeast(out)

% Tile 2
nexttile
plotbeast(out)