Store values from loop

Question

Simon Christensen 2024-3-8

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2092101-store-values-from-loop

评论： Rik 2024-3-8

Hi Matlab,

I'm running a program, calculating the recurvie residuals (see code below). The loop indicates the model is run for some base period (Z) until end of the sample.

for (i = Z:206)

%data

vX = [LOG_GDP(1:i,:,1), gvm_GDP(1:i,:,1), goverment_bond(1:i,:,1), t_bill(1:i,:,1), inflation(1:i,:,1), LOG_private_investment(1:i,:,1), LOG_mon_base(1:i,:,1), LOG_gvm_spending(1:i,:,1), excess_return(1:i,:,1), Unemployment(1:i,:,1), LOG_private_con(1:i,:,1), LOG_house(1:i,:,1)];

%function to calculate residuals from VECM

[res1] = f(vX, K, r, p)

end

However, I do not understand how to store the estimates forthe loop. If I run the program, it only returns the values for last value.

Does someone know how to store the values?

Any help is greatly appreciated

Best regards

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Rik 2024-3-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2092101-store-values-from-loop#answer_1422781

在 MATLAB Online 中打开

If you want to store results from all iterations, you need to store them and not overwrite them:

res1 = cell(1,100);index = 0;
for i = Z:206
    %data
    vX = [LOG_GDP(1:i,:,1), gvm_GDP(1:i,:,1), goverment_bond(1:i,:,1), t_bill(1:i,:,1), inflation(1:i,:,1), LOG_private_investment(1:i,:,1), LOG_mon_base(1:i,:,1), LOG_gvm_spending(1:i,:,1), excess_return(1:i,:,1), Unemployment(1:i,:,1), LOG_private_con(1:i,:,1), LOG_house(1:i,:,1)];
    %function to calculate residuals from VECM
    index = index+1;
    res1{index} =  f(vX, K, r, p);
end
res1((index+1):end) = []; % remove unused elements

3 个评论
显示 1更早的评论隐藏 1更早的评论

Simon Christensen 2024-3-8

Actually, I think i found the solution to the above by myself: It is just to rearrange the function outout. All good!

Rik 2024-3-8

在 MATLAB Online 中打开

You should not use numbered variables. That encodes data in the variable name, which is a fragile method of design that doesn't scale well.

res = cell(100,4);index = 0;
for n=1:3
    index = index+1;
    [res{index,:}]=example_function;
end
res((index+1):end,:) = []; % remove unused elements
disp(res)
    {[1.1713]}    {[2.3546]}    {[3.6638]}    {[4.5237]}
    {[1.4065]}    {[2.1759]}    {[3.1997]}    {[4.8497]}
    {[1.1356]}    {[2.5412]}    {[3.4620]}    {[4.7298]}
% example function with 4 outputs
function [a,b,c,d]=example_function
a=1+rand;b=2+rand;c=3+rand;d=4+rand;
end