Combine arrays of different dimensions

Hello,

If I have 2 data sets of different dimensions.

xdata1 = [1 2 3 4] ; ydata1 = [2 3 4 5]
xdata2 = [1 2 3 4 5]; ydata2 = [7 4 9 11 12]

Can I create a combined data set for xdata and ydata such that:

xdata = [1 2 3 4; 1 2 3 4 5]
ydata = [2 3 4 5; 7 4 9 11 12];

I tried with initializing an array of zeros or NaN for xdata and ydata and then replace with xdata1 and ydata1 elements. But I do not want zeros or NaN in the array as I am using these numbers for further calculation which gives error with zero or NaN elements. I basically need to store all the xdatas and ydatas together and then call each colum (representing each data set) individually later for calculation.

Can you help?

Answer 1

[Adam Danz]

You cannot have a matrix with varying row lenght.

One alternative is to create a padded array where NaN values fill the ends of shorter rows.

xdata = [1 2 3 4 NaN; 1 2 3 4 5];
ydata = [2 3 4 5 NaN; 7 4 9 11 12];

"I do not want zeros or NaN in the array as I am using these numbers for further calculation which gives error with zero or NaN elements."

To deal with that, many Matlab functions automatically ignore NaN values or have flags that can be set to ignore NaN values. For the functions that do not have this option, you can use isnan() to identify non-NaN values so that NaNs are ignored.

mean(xdata,2,'Omitnan')

An alternative is to pack the vectors into a cell array.

xdata = {[1 2 3 4]; [1 2 3 4 5]};
ydata = {[2 3 4 5]; [7 4 9 11 12]};

and to use cellfun() to process each element of the array.

cellfun(@(x)mean(x,'omitnan'),xdata)

[addendum]

"I basically need to store all the xdatas and ydatas together and then call each column"

I think the first method (padding) would work best.

[Nimasha Pilippange]

I have 2 matrices as A=[ 1 2 3 4 5] and B= [ 6 7 8 9 10] , How can I combine them so I can get C=[1 6 2 7 3 8 4 9 5 10]. I need for a large matrix. But first I will start with a small matrix.

Can anyone help?

[Adam Danz]

It looks like you found an answer to your question.