Storing regression coefficients in a loop
3 次查看(过去 30 天)
显示 更早的评论
Hello,
I'm doing a simple linear regression on a 20x30x365 matrix (latitude, longitude, time). I want to store the slope and p - value for each lat/lon pair such that the end result are two 20x30 matrices. I'm attaching some dummy data and my attempted code below. Thank you.
X = size(prec,3);
i = 20;
j = 30;
slopes = zeros(i,j);
pvals = zeros(i,j);
for i = 1:i
for j= 1:j
mdl = fitlm(X,prec(i,j,:));
slope = table2array(mdl.Coefficients(2,1));
pval = table2array(mdl.Coefficients(2,4));
slopes = slope(i,j);
pvals = pval(i,j);
end
end
0 个评论
采纳的回答
VBBV
2021-12-7
编辑:VBBV
2021-12-8
slopes(i,j) = slope;
pvals(i,j) = pval;
2 个评论
VBBV
2021-12-8
n = size(prec,3);
X = (1:1:n)';
I = 20; % change this variable with a differnt letter
J = 30; % same here
slopes = zeros(I,J);
pvals = zeros(I,J);
for i = 1:I
for j= 1:J
y = reshape(prec(i,j,:),[n,1]);
mdl = fitlm(X,y);
slope = table2array(mdl.Coefficients(2,1));
pval = table2array(mdl.Coefficients(2,4));
slopes(i,j) = slope;
pvals(i,j) = pval;
end
end
When the for loop runs, its using concurrent values of i and j. Change the symbols as above and run it
更多回答(0 个)
另请参阅
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!