Confidence intervals in Contour Plots

4 次查看（过去 30 天）
David Canales
David Canales 2017-6-7
编辑： David Canales 2017-6-8
I am trying to generate a contour plot that gives me the 95% and 68% interval probability. What I should obtain is the following:
where the inner circle is the 68% and the outer 95%. I am trying to do the following with the input CODAchain1.txt: the x values are from row 0-1000, and y from row 1001-2000.
I am trying to do that with R, but I want to resort to Matlab again for my research and I want to see its abilities for statistics. The code I would use is the following:
ic_table_file1 = 'CODAchain1.txt';
%%Read tables
codachain = load(ic_table_file1);
%%Length params
snum=1000;
long_vec=length(codachain);
%%Computations mean
media=mean(codachain(1:snum,2));
mean2=mean(codachain(snum+1:long_vec,2));
%Standard deviation
sdev=std(codachain(1:snum,2));
sdev2=std(codachain(snum+1:long_vec,2));
%Percentage 68
percentage=0.68;
frac=int64(percentage*snum);
vector1=codachain(1:snum,2);
vector1=sort(vector1);
i_fin=snum-frac+1;
for i=1:1:i_fin
vector_res(i,1)=vector1(frac+i-1)-vector1(i);
end
minres=min(vector_res);
for i=1:1:i_fin
if (vector_res(i,1)==minres)
k_res=i;
end
end
int_min68=vector1(k_res,1);
int_max68=vector1(frac+k_res-1,1);
%Percentage 95
percentage=0.9;
frac95=percentage*snum;
i_fin95=snum-frac95+1;
for i=1:1:i_fin95
vector_res95(i,1)=vector1(frac95+i-1)-vector1(i);
end
minres95=min(vector_res95);
for i=1:1:i_fin95
if (vector_res95(i,1)==minres95)
k_res95=i;
end
end
int_min95=vector1(k_res95,1);
int_max95=vector1(frac95+k_res95-1,1);
%Contour plots
figure(3);
hold on;
k68=0;
for i=1:1:snum
if(codachain(i,2)>int_min68&&codachain(i,2)<int_max68)
k68=k68+1;
codachain68x(k68,:)=codachain(i,:);
codachain68y(k68,:)=codachain(i+snum,:);
end
end
k95=0;
for i=1:1:snum
if(codachain(i,2)>int_min95&&codachain(i,2)<int_max95)
k95=k95+1;
codachain95x(k95,:)=codachain(i,:);
codachain95y(k95,:)=codachain(i+snum,:);
end
end
x95=codachain95x(1:k95,2);
y95=codachain95y(1:k95,2);
x68=codachain68x(1:k68,2);
y68=codachain68y(1:k68,2);
xout=codachain(1:snum,2);
yout=codachain(snum+1:long_vec,2);
plot(xout,yout,'.r');
So here I obtain the distribution of the data.
Now what I would need to do is to obtain the 68% and 95% interval so that I obtain the same plot as shown in the beginning.
Thank you very much.

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Contour Plots 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by