Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Browse files
Browse the repository at this point in the history
Add files via upload
- Loading branch information
Showing
2 changed files
with
214 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,214 @@ | ||
num=xlsread('muon.xlsx'); | ||
x=(num(:,1)); | ||
x=x./1000; | ||
step=0.4; | ||
X=0:step:max(x); | ||
figure(1) | ||
h=histogram(x,X); | ||
xlabel('Decay Time (us)') | ||
ylabel('Observed Counts') | ||
hold on | ||
y=h.Values; | ||
XX=(step/2):step:max(x); | ||
XX=XX(1:end-1); | ||
errorbar(XX,y,sqrt(y),'o') | ||
|
||
|
||
%linear and nonlinear fit before background subtraction | ||
%plot(XX,log(y),'o') | ||
fun1=@(v) v(1)*exp(-v(2)*XX)-y; | ||
x0=[4000,1]; | ||
[v,resnorm,residual,exitflag,output,lambda,jacobian]=lsqnonlin(fun1,x0); | ||
|
||
|
||
Jacobian = full(jacobian); | ||
varv=resnorm*inv(Jacobian'*Jacobian)/length(XX); | ||
stdv=sqrt(diag(varv)); | ||
|
||
|
||
Y1=v(1)*exp(-v(2)*XX); | ||
plot(XX,Y1,'g') | ||
legend('Observed Decays','Uncertainty','Best Fit Without Background Subtraction') | ||
title('Nonlinear Regression with no Consideration of Bacground Effects') | ||
ylabel('Bin Count') | ||
xlabel('Decay Time (microseconds)') | ||
%hold on | ||
%plot(XX,y) | ||
|
||
|
||
|
||
%% | ||
%linear and nonlinear fit after background subtraction | ||
clearvars v resnorm residual stdv jacobian | ||
|
||
figure(2) | ||
count=sum(y(XX>12)); | ||
avg=count/(length(y(XX>12))); | ||
Y=y-avg; | ||
Y(Y<0)=0; | ||
%plot(XX,log(Y),'o') | ||
h=histogram(x,X); | ||
xlabel('Decay Time (us)') | ||
ylabel('Observed Counts') | ||
hold on | ||
y=h.Values; | ||
errorbar(XX,y,sqrt(y),'o') | ||
|
||
|
||
|
||
|
||
%fun2=@(v) v(1)*exp(-v(2)*XX)+v(3);%-y; | ||
%v=[50,5,1]; | ||
%fun2=@(v,x) (v(1)*exp(-v(2)*x)+v(3)); | ||
%nlinfit(XX,y,fun2,X0) | ||
|
||
fun2=@(v) v(1)*exp(-v(2)*XX)+v(3)-y; | ||
X0=[4000,5,10]; | ||
|
||
[v,resnorm,residual,exitflag,output,lambda,jacobian]=lsqnonlin(fun2,X0); | ||
ci=nlparci(v,residual,'Jacobian',jacobian); | ||
v=v'; | ||
|
||
|
||
Jacobian = full(jacobian); | ||
varv=resnorm*inv(Jacobian'*Jacobian)/length(XX); | ||
stdv=sqrt(diag(varv)); | ||
|
||
|
||
|
||
|
||
|
||
Y=v(1)*exp(-v(2)*XX)+v(3); | ||
YY=log(v(1))-v(2)*XX+log(v(3)); | ||
plot(XX,Y,'g') | ||
legend('Observed Decays','Uncertainty','Best Fit With Background Subtraction') | ||
title('Nonlinear Regression with Consideration of Bacground Effects') | ||
ylabel('Bin Count') | ||
xlabel('Decay Time (microseconds)') | ||
%hold on | ||
%plot(XX,y) | ||
|
||
|
||
|
||
|
||
|
||
%% Maximum likelihood | ||
%without background subtraction | ||
lambda=0.275:0.0001:0.282; | ||
%lambda=0.1:0.01:0.6; | ||
P=zeros(length(lambda),length(x)); | ||
mP=zeros(1,length(lambda)); | ||
for i=1:length(lambda) | ||
for j=1:length(x) | ||
P(i,j)=lambda(i)*exp(-lambda(i)*x(j)); | ||
end | ||
end | ||
P=log(P); | ||
|
||
for i=1:length(lambda) | ||
mP(i)=sum(P(i,:)); | ||
end | ||
|
||
figure(5) | ||
plot(lambda,mP) | ||
hold on | ||
yy=(max(mP(:))-0.5)*ones(1,length(lambda)); | ||
plot(lambda,yy) | ||
vs=(mP((mP-yy>-0.04 & mP-yy<0.04))); | ||
Vs=zeros(1,2); | ||
for i=1:length(vs) | ||
Vs(i)=lambda(mP==vs(i)); | ||
end | ||
plot(Vs,yy(1:2),'o') | ||
|
||
|
||
lambda1=lambda(mP==max(mP)); | ||
siga=abs(lambda1-Vs(1)); | ||
sigb=abs(lambda1-Vs(2)); | ||
yy1=get(gca,'ylim'); | ||
plot([lambda1,lambda1],yy1); | ||
title('Maximum Likelihood Approach with no Consideration of Background') | ||
xlabel('Lambda (s^-1)') | ||
ylabel('Sum of log of Probabilities') | ||
legend('Sum of log of Probabilities','Maximum-0.5') | ||
|
||
|
||
|
||
%% with background subtraction | ||
lambda=0.48:0.0002:0.50; | ||
b=0.0095:0.00002:0.0101; | ||
|
||
P=zeros(length(lambda),length(b),length(x)); | ||
mP=zeros(length(lambda),length(b)); | ||
for i=1:length(lambda) | ||
for j=1:length(b) | ||
for k=1:length(x) | ||
P(i,j,k)=lambda(i)*(1-b(j)*20)*exp(-lambda(i)*x(k))+b(j); | ||
end | ||
end | ||
end | ||
P=log(P); | ||
|
||
for i=1:length(lambda) | ||
for j=1:length(b) | ||
mP(i,j)=sum(P(i,j,:)); | ||
end | ||
end | ||
|
||
figure(6) | ||
[XXX,YYY]=meshgrid(b,lambda); | ||
surf(XXX,YYY,mP) | ||
xlabel('Background') | ||
ylabel('Lambda (s^-1)') | ||
zlabel('Sum of log of Probabilities') | ||
hold on | ||
|
||
M=max(mP(:)); | ||
[row,col]=find(mP==M); | ||
|
||
|
||
lambda2=lambda(row); | ||
background=b(col); | ||
|
||
|
||
%% Uncertainty | ||
L=M-0.5; | ||
[row,col]=find(mP>L-0.03 & mP<L+0.03); | ||
xp=b(col); | ||
yp=lambda(row); | ||
zp=L*ones(length(row),length(row)); | ||
xp=xp'; yp=yp'; | ||
k=boundary(xp,yp); | ||
LL=length(k); | ||
zp2=L*ones(LL,1); | ||
plot3(xp(k),yp(k),zp2,'r','Linewidth',2) | ||
title('Maximum Likelihood Approach with Consideration of Background') | ||
hold off | ||
|
||
|
||
figure(7) | ||
contourf(YYY,XXX,mP) | ||
hold on | ||
scatter(yp,xp,'b') | ||
xlabel('Lambda (s^-1)') | ||
ylabel('Background') | ||
zlabel('Sum of log of Probabilities') | ||
|
||
|
||
yy1=get(gca,'ylim'); | ||
h1=plot([lambda2,lambda2],yy1); | ||
sig1=abs(lambda2-min(yp)); | ||
sig2=abs(lambda2-max(yp)); | ||
|
||
ul=xp(yp==max(yp)); | ||
ll=xp(yp==min(yp)); | ||
|
||
h2=plot([lambda2-sig1,lambda2],[ll(1),ll(1)]); | ||
|
||
h3=plot([lambda2+sig2,lambda2],[ul(3),ul(3)]); | ||
%legend('','''Value of Lambda','Sigma 1','Sigma 2') | ||
legend([h1 h2 h3],{'Value of Lambda Determined','Sigma (-)','Sigma (+)'}) | ||
title('Uncertainties in the Multidimensional Maximum Likelihood Approach') | ||
|
||
|
||
|