Storing data for "for loop"

Hi everyone,

for my L0 = [1 2 3 4 5], L10 = [2 3 4 5 6] L20 = [0 9 2 3 4] .... L200 = [2 3 4 5 6] (Up to 20 L values, just type random L for asking)

and x = [0:1:4]

I want to perform a polyfit function for 6th polynomial order for each L, with same x, but i do not want to each time type "polyfit(x,L0,6)" and then "polyfit(x,L20,6)

My question is : How can I make a for loop for this

I have tried L = [L0 L10 L20 ... L200],

but this way, it will store all values in one vector.

Please help ASAP !!

Thanks in advance

Answer 1

[Star Strider]

Try something like this:

L0 = [1 2 3 4 5];
L10 = [2 3 4 5 6];
L20 = [0 9 2 3 4];
L = [L0; L10; L20];
x = 0:4;

n = size(L,2)-2;
for k = 1:size(L,1)
    p(k,:) = polyfit(x,L(k,:),n);
end

This creates a matrix of row vectors in ‘L’. Note that I set ‘n’ to be 2 less than the vector lengths. A 6-order polynomial might be appropriate for your actual data, although it will crash here. (Please re-consider fitting a 6-order polynomial anyway.)

[Khang Nguyen]

Thanks you so much. It works :)).

[Star Strider]

As always, my pleasure!

Answer 2

[Josh]

You can store your L variables either as rows in a matrix:

% Create L matrix with a different L value in each row (I only put in three rows for simplicity)
L = [1, 2, 3, 4, 5; 2, 3, 4, 5, 6; 0, 9, 2, 3, 4]; 

% Create x value
x = 0:4;

% Create a result matrix; this will store the output of polyfit as separate rows:
order = 6;
results = zeros(size(L, 1), order + 1);

% Calculate the results
for i = 1 : size(L, 1)
    results(i, :) = polyfit(x, L(i, :), order);
end

% The syntax L(i, :) returns the entire ith row of the matrix

[Khang Nguyen]

I have tried your code, and it's only assign the value of the last L to the result matrices

Answer 3

[Geoff Hayes]

Khang - don't create variables just for the sake of having variables. If all of your L arrays are of the same dimension, then just store them in a (for example) 20x5 matrix where each row corresponds to one of your (no longer needed) L variables. You would then iterate over each row and call polyfit on that row. For example,

for k=1:size(myData,1)
    [p,S,mu] = polyfit(x,myData(k,:),6);
    % store the results in an appropriately sized output matrix
end

where myData is your 20x5 array. If not all L arrays are of the same dimension, then just store all in a cell array.