I have a set of data (say 40) each having three Input values Xi, Yi and Zi. I also have the corresponding output values Xo, Yo and Zo for these data. So the input and out values have 40 X 3 vectors. I would like to know what is the best method for minimizing error so that the new output values become closer to the input values.
I think the method depends on the nature of the system through which you pass your inputs through.
For example if the nature of the system is fairly linear the you can use a simple averaging out menthod....etc.
~Kalyan.
One way to find the error is to caculate the norm of actuall data and its corresponsing output and try to minimize the norm value in order to minimize the error.