Welcome to exchange and study!
Using Matlab can also draw a beautiful confidence interval and normal distribution Curve curve. Here is the specific code. It's worth noting that the following code is based on Matlab 2018b.
%% ================================================== % Author: Zephyr Hou % Time: 2020-09-26 %% Main Function clc;clear;close all; x=-4:0.01:4; y=gaussmf(x,[1,0]); plot(x,y,'b','LineWidth',2);hold on; axis([-4 4 0 1.001]) %% 置信区间边界 plot([-2,-2],[0,gaussmf(-2,[1,0])],'k','LineWidth',2);hold on plot([2,2],[0,gaussmf(2,[1,0])],'k','LineWidth',2);hold on %% 置信区间颜色填充 for i=1:floor(length(x)/4) plot([x(i),x(i)],[0,gaussmf(x(i),[1,0])],'Color',[0.35,0,0.6],'LineWidth',2);hold on end for i=1:floor(length(x)/4) ind=x(x>2); plot([ind(i),ind(i)],[0,gaussmf(ind(i),[1,0])],'Color',[0.35,0,0.6],'LineWidth',2);hold on end %% 图中标注设置 set(gca,'XTicklabel',{' ',' ','\mu-2\sigma',' ','\mu',' ','\mu+2\sigma',' '},'fontsize',15) set(gca,'YTicklabel',[]) annotation('arrow',[0.27,0.30],[0.24,0.20],'LineWidth',2) annotation('arrow',[0.76,0.73],[0.24,0.20],'LineWidth',2) text(-3,0.20,'\alpha/2=0.025','color','r','fontsize',15) text(2.1,0.20,'\alpha/2=0.025','color','r','fontsize',15) title('The 95% confidence interval for \mu','fontsize',15) %% ===================================================
The operation results is as follows:
Reference:
If you use this program to draw a normal distribution curve, please cite the following paper.