ex6_ch3.m

% Chrysa Tsimperi
% Data Analysis 2021
% Chapter 3 Excerise 6
% on bootstrap estimate of standard error of sample mean
n = 10;
mu = 0;
sigma = 1;
B = 1000;
doexponential = 0; % If 1 do exponential transform first

% rng(1);
xV = mu+sigma*randn(n,1);
% (c) If doexponential=1 do exponential transform first
if doexponential
    xV = exp(xV);
end
mx = mean(xV);

bootmxV = NaN(B,1);
for iB=1:B
    rV = unidrnd(n,n,1);
    xbV = xV(rV);
    bootmxV(iB) = mean(xbV);
end

% (a) Draw the histogram of bootstrap statistics.
figure(1)
clf
hist(bootmxV)
hold on
yaxV = ylim;
plot(mx*[1 1],yaxV,'r')
xlabel('$\bar{x}$','Interpreter','latex')
ylabel('empirical f of $\bar{x}$','Interpreter','latex')
title(sprintf('B=%d bootstrap means for sample of size n=%d',B,n))

% (b) Bootstrap estimate of standard error of sample mean
parsemu = std(xV)/sqrt(n);
bootsemu = std(bootmxV);
fprintf('Estimate of standard error of sample mean = %1.3f \n', parsemu);
fprintf('Estimate of standard error of sample mean = %1.3f \n', bootsemu);