How to fit an exponential curve for every column in a matrix?

I have a matrix of 257x36 and each column represents a different data set.

I need to apply an exponential fit to each column and also store the fit values.

How can I go about doing this?

[Torsten]

And what are the corresponding x-values to the y-values in the columns ?

[parslee]

x is 1 to 257 for all of them

Answer 1

[David Hill]

for k=1:36
  B{k}=fit((1:257)',A(:,k),'exp1');
end

[parslee]

Is there a way to extract the a and b values and store them in a matrix?

[Torsten]

for k=1:36
    B{k} = fit((1:257)',A(:,k),'exp1');
    COEFFS(k,:) = coeffvalues(B{k});
end

[parslee]

I want to store all this in another matrix called y where it stores y1, y2, y3, and so on using the exponential fit: a * exp(b*x).

I tried writing a for loop and it's giving me an error. It says fittype objects can't be assigned to using subscripts.

for k = 1:36
    y{k} = coeffs(i,1) * exp(coeffs(i,2)*x);
end

I'm not sure what I'm doing wrong.

[Torsten]

for k=1:36
    B{k} = fit((1:257)',A(:,k),'exp1');
    coeffs = coeffvalues(B{k});
    Ahat(:,k) = coeffs(1)*exp(coeffs(2)*(1:257).');
end

[parslee]

Thank you!!