Adding text from calculated functions and line of best fit to scatter plot

I have calculated three statistic values for observed vs simulated data. I want to add these defined output values (i.e. the cor_Manual_Urbano, bias_da_Manoel_Urbano, NASH_Manoel_Urbano) to the scatter plot of oberserved vs simulated. There is the additional problem that the output result for cor_Manual_Urbano produces a matrix but I only want to display the correlation value between the two variables. I also want to add a line of best fit to the scatter plot. Here is my code so far:

%pearsons%
cor_Manoel_Urbano = corrcoef(combined_data.Observed_Manoel_Urbano,combined_data.Simulated_Manoel_Urbano, 'Rows', 'pairwise')
%bias%
just_Manoel_Urbano = combined_data(:, [11,22]);
cleaned_Manoel_Urbano = rmmissing(just_Manoel_Urbano);
observedmean_Manoel_Urbano = mean(cleaned_Manoel_Urbano.Observed_Manoel_Urbano);
simulatedmean_Manoel_Urbano = mean(cleaned_Manoel_Urbano.Simulated_Manoel_Urbano);
bias_da_Manoel_Urbano = 100*((observedmean_Manoel_Urbano-simulatedmean_Manoel_Urbano)/observedmean_Manoel_Urbano)
%NASH%
DATA_NASH_Manoel_Urbano = [combined_data.Observed_Manoel_Urbano,combined_data.Simulated_Manoel_Urbano];
DATA_NASH_Manoel_Urbano(any(isnan(DATA_NASH_Manoel_Urbano), 2), :) = [];
NASH_Manoel_Urbano = NSE(DATA_NASH_Manoel_Urbano(:,1), DATA_NASH_Manoel_Urbano(:,2))
%scatter%
figure22 = figure('Name','Scatter Manoel Urbano','Color',[1 1 1]);
scatter(cleaned_Manoel_Urbano.Observed_Manoel_Urbano, cleaned_Manoel_Urbano.Simulated_Manoel_Urbano, 'k')
xlabel('Observed','FontName','Calibri','FontSize',12)
ylabel('Simulated', 'FontName','Calibri','FontSize',12)
title('Manoel Urbano','FontName','Calibri','FontSize',14);
set(gcf,'Position',[100 100 400 400])

[Elizabeth Lees]

Still looking for a solution to adding a line of best fit to the scatter() function

Answer 1

[Star Strider]

To display the information on the plot, use the text function.

The sprintf or compose functions can also be helpful, depending on what you want to do.

[Elizabeth Lees]

using the text function with a predefined variable e.g. NASH_Manoel_Urbano gives me an error of 'Error using text Character vector argument expected after 2 or 3 numeric arguments'

this is my code for reference:

text(200,800,NASH_Aruma_Jusante)

[Star Strider]

I am not certain what ‘NASH_Aruma_Jusante’ is.

I suggest using sprintf (that I linked to in my Answer) to print it if ti is not already a character array or a string variable.

[Elizabeth Lees]

Sprintf still doesn't help with the problem that corrcoef() function gives an output of a 2x2 matrix therefore when using sprintf() I'm getting all four numbers without spacing when I only want the second value

[Star Strider]

Try this:

text(200,800,sprintf('Correlation = %.2f',NASH_Aruma_Jusante(2)))

Change the subscript (here (2)) to display the value you want. I have no idea which one it is, so you will have to experiment.

[Elizabeth Lees]

Thank you!!! Works amazingly!

[Star Strider]

My pleasure!

[Elizabeth Lees]

Do you know a solution for adding line of best line for the scatter() function?

[Star Strider]

Yes!

If you have the Statistics and Machine Learning Toolbox, use the lsline function. Otherwise, use the ployfit and polyval functions.

[Elizabeth Lees]

lsline function works brilliantly! thank you so much all for your help!

[Star Strider]

As always, my pleasure!