CODEWITHSUNDEEP
matlabmatrix
metonymic
metonymic

Reputation: 1

Error using / Matrix dimensions must agree

I'm attempting to create a Lorentzian distribution for my data using an anonymous function. The Gaussian distribution is created with no issues, but the Lorentzian distribution encounters this error.

The error:

Error using /

Matrix dimensions must agree.

Error in @(p,x)1/(pi*p(2)*(1+((x-p(1))/(p(2))).^2))

Code:

% Variables necessary for histograms
num_bins = 100;
low_range = linspace(min(low_data), max(low_data), num_bins);
low_counts = histc(low_data, low_range);
low_err = sqrt(low_counts);  


% Creating and testing fits
gauss_fun_form = @(p, x) 1/(p(2)*sqrt(2*pi))*exp(-(x-p(1)).^2/(2*p(2).^2));
loren_fun_form = @(p, x) 1/(pi*p(2)*(1+((x-p(1))/(p(2))).^2));

low_avg = mean(low_data);
low_std = std(low_data);    

low_gauss_param = [low_avg low_std];
low_gauss = gauss_fun_form(low_gauss_param, low_range);
low_gauss_scale = max(low_counts)/max(low_gauss);

low_loren_param = [low_avg 2*log(2)*low_std];
low_loren = loren_fun_form(low_loren_param, low_range);
low_loren_scale = max(low_counts)/max(low_loren);

Upvotes: 0

Views: 167

Answers (1)

michaeltang
michaeltang

Reputation: 2898

(pi*p(2)*(1+((x-p(1))/(p(2))).^2)) generate vector of 100 dimensions

while 1 is vector of 1 dimension, so matrix division is not valid , ./ should be applied. change the definition of loren_fun_form as below should work

loren_fun_form = @(p, x) 1./(pi*p(2)*(1+((x-p(1))/(p(2))).^2));

Upvotes: 1

Related Questions