How do replace missing data in a matrix with interpolated values

Hello,

I'm trying to replace missing data within a matrix with interpolated values. For example, if I have a matrix with a few integer values and many missing data points like below.

data(1:10,1:10) = nan;
data(2,3) = 2; data(5,6)=3; data(8,1) = 1; data(1,8) = 4; data(8,8) = 1; data(5,9) = 2;data(5,7)= 2;

How can I replace the NaN entries with interpolated values? I have tried using the fillmissing function, but it seems to just go along rows or columns.

Thanks so much for your time!

Colby

Answer 1

[Image Analyst]

Colby, well that's not much data to go on. You might try kriging (Click here) a geospatial estimation method they use to, for example, predict temperatures and rainfall everywhere when they actually only have measurements at certain isolated locations (where weather stations are located).

There are kriging programs in the File Exchange: http://www.mathworks.com/matlabcentral/fileexchange/?utf8=%E2%9C%93&term=kriging