Iterating over a vector to create a cell array without for loop

I would like to know if it is possible to do the same as the code below without a for loop:

for i=1:length(index)
    date1{i}=datenum(datos.Date(index(i):end));
    Y{i}=(datos.High(index(i):end)+datos.Low(index(i):end))./2;
    X{i}=[ones(length(Y{i}),1) date1{i}];
    b{i}=regress(Y{i},X{i});
    tendency{i}=b{i}(1)+b{i}(2)*X{i}(:,2);
    parallel{i}=std(Y{i});
end

[Abhishek Gupta]

It would be helpful, if you could share your data containing 'datos'.

Answer 1

[Jan]

Most likely this cannot be vectorized. By why do you ask for it? Do you want to accelerate the code? This is possible:

Avoid repeated work. E.g.

date1{i}=datenum(datos.Date(index(i):end));

Convetry many date repeatedly. So move this befor the loop to convert the dates once only:

date1V = datenum(datos.Date);
% and inside the loop:
  date1{i} = date1V(index(i):end);
  

regress() is a very powerful function. All you want to do is a polynomial fit. Then this is probably faster:

function p = TinyPolyFit(x, y, n)
% Construct Vandermonde matrix:
x = x(:);
V = ones(length(x), n + 1);
for j = n:-1:1
   V(:, j) = V(:, j + 1) .* x;
end

% Solve least squares problem:
[Q, R] = qr(V, 0);
p      = transpose(R \ (transpose(Q) * y(:)));
end

[Carlos de los Santos]

Thank you so much. I did want to accelerate the code and it worked perfectly.