maximize log likelihood function

Jun Yoo

2020 8 28

0 个回答

5 次查看（30 天）

0 个投票

Hi, Ive been stuggling with this code for maximizing log likelihoood.

I recently started to learn Matlab so Im confused with function in Matlab and how it works.

hope I can get some help from here

here is my code.

data = readtable('MROZ.csv');
 
data = data{:,[1, 20, 6, 19, 22, 5, 3, 4]};
 
y = data(:,1);
x = data(:,[2,3,4,5,6,7,8]);
int = ones(753, 1);
X = horzcat(int, x);
beta = X\y; % LPM coefficient; 
 
[beta,fval,exitflag] = fminunc(@probitnll,beta,X,y);
 
function nll = probitnll(beta,X,y)
 
  yhat = X * beta;  
  
  G = normcdf(yhat);
  
  nll = sum(y .* log(G) + (1-y) .* log(1-G));
  
end