How to extract the values of x and y from this matlab regression models?

I have executed the matlab regression modeling (see the attached screenshot) but i want to extract the data of x,y into excel sheet to use it in drawing other plots.

can any one help me?

Regards

Ahmed

[dpb]

Use the "export model" and read about fit postprocessing to use the resulting fit object.

[Adam Danz]

What do you mean by "the data of x,y"? Do you mean you want to access the (x,y) coordinates that are plotted within the Regression Learner App?

There might be a better way to do this but you could extract them directly from the figure.

1. Get the handle to the app's axes. Here are two ways to do that.
2. Once you have the handle, let's call is RLax, you can get the (x,y) coordinates of each plotted object using two lines below.

X = get(RLax.Children, 'XData'); 
Y = get(RLax.Children, 'YData');

X and Y will be cell arrays of vectors, one element for each line object.

[Ahmed Elbeltagi]

Thanks a lot for your help Mr. Adam Danz Your help

I have applied your steps and reached to this point as shown in attached screenshot.

How to extract values of Y{1,1} and Y{2,1} into excel sheet. OR, How to extract the values of Y into excel sheet file.

Waiting for your kind reply.

Regards

[Adam Danz]

Glad it's working out. I copied the comment to the answers section and addressed your additional questions.

Answer 1

[Adam Danz]

Step 1) Access the (x,y) coordinates that are plotted within the Regression Learner App. If you can't extract the data directly from "export model", you could extract them from the app's figure.

1. Get the handle to the app's axes. Here are two ways to do that.
2. Once you have the handle, let's call is RLax, you can get the (x,y) coordinates of each plotted object using two lines below.

X = get(RLax.Children, 'XData'); 
Y = get(RLax.Children, 'YData');

X and Y will be cell arrays, one element for each line object, containing 1*n vectors.

Step 2) Convert the cell arrays to matricies. If each vector within the cell array is the same length, convert the cell array to a matrix using,

m = cell2mat(Y')';  

If the vectors contain a different number of elements, you'll need to pad the shorter vectors with NaN values using,

maxNumCol = max(cellfun(@numel, Y));  % max length
m = cell2mat(cellfun(@(c){padarray(c,[0,maxNumCol-size(c,2)],NaN,'Post')},Y))';

Both versions result in an m*n matrix of m observations and n variables.

Step 3) Write the data to an excel file. If you only want to write the matrix,

writematrix(m, 'myData.xlsx')

If you want to include column headers Y1, Y2, ..., Yn

T = array2table(m, 'VariableNames', compose('Y%d',1:size(mPad,2)));
writetable(T, 'myData.xlsx')

For more info on writing to excel:

https://www.mathworks.com/help/matlab/import_export/exporting-to-excel-spreadsheets.html

[Ahmed Elbeltagi]

Done Mr. Adam Danz

Thanks a lot for your help

Many thanks again.

Cheers

[Adam Danz]

Glad I could help.