How do I merge 3 estimated parameters stored in 3 different cells?

If I use

load para1
m1=para;
load para2
m2=para;
load para3
m3=para;
m=merge(m1,m2,m3)

it gives the error Undefined function 'merge' for input arguments of type 'cell' Para is the cell 20X1 where the parameters are stored.

Kindly help, Thanks in advance.

Answer 1

[Star Strider]

The cat function may be what you’re looking for.

[Tejaswini Ramesh]

I do not want them concatenated but merging of the parameters as in, parameter vector is a statistically weighted mean (using the covariance matrices to determine the weights) of the parameters of para1, para2 and para3.

[Stephen23]

@Tejaswini Ramesh: Neither the word "covariance" nor "mean" appears anywhere in your question. Sorry, but we can't read minds yet, although we do try.

Can you please edit your question and instead of giving us some broken code that uses a non-existent function you should actually tell us details of what you want to do. This includes examples of the input and output values.

Please read this, and follow its advice if you actually want to get some help:

http://www.mathworks.com/matlabcentral/answers/6200-tutorial-how-to-ask-a-question-on-answers-and-get-a-fast-answer

[Star Strider]

Do you have some sort of reference for that? I never heard of that procedure.

If you want to do inverse-variance weighting and then add the weighted parameters, with the variances of the parameters being the corresponding elements of the diagonal of the covariance matrix, something like this may work:

m = para1./diag(cov1) + para2./diag(cov2) + para3./diag(cov3);

where ‘cov1’ is the covariance matrix associated with ‘para1’, and so for the others.

I make no claim that this is a statistically valid technique.