How to optimize the value of x(2)

3 次查看（过去 30 天）
Raj Arora
Raj Arora 2022-5-2
编辑： Torsten 2022-5-2
CODE FOR COMPUTING MEDIAN AND MAXIMUM LIKELIHOOD VALUE. I WANT TO CALCULATE OPTIMUM VALUE OF STANDARD DEVIATION BASED ON INOUT DATA FILE
function likeli
da = load("C:\Users\Admin\OneDrive - IIT Bombay\Desktop\standard deviation\fragility1.txt")%%Damage state in terms of 0 and 1
WH = load("C:\Users\Admin\OneDrive - IIT Bombay\Desktop\standard deviation\velocity1.txt")%%
n = length(da)
xdamage = da(:,1)
x0 = [2]%x valueis a median values%
options = optimset('LargeScale','off','Display','off','TolX',0.001,'TolFun',0.001)
[x,fval] = fminsearch(@myfun,x0,options,n,WH,xdamage)%fval is the likeli hood function%
function f = myfun(x,n,WH,xdamage);
options = optimset('LargeScale','on','Display','off','TolX',0.001,'TolFun',0.001)
p1=0.0;
for i=1:n
x(2) = 0.5;
yx=(log(WH(i)/x(1)))/x(2)%%x(2) is standard deviation
if yx >= 5.0;
y1=5.0;%%maximum value normcdf can take taken as 5%%
elseif yx<=-5.0;
y1=-5.0;
else
y1 = yx;
end
y2=normcdf(y1)
p1=p1+log(((y2)^xdamage(i))*((1.0-y2)^(1.0-xdamage(i))))%%Maximum likeli hood%
end
f=-p1;
return
Hello all I have a querry, I am solving one problem in which I have to compute optimum value of x(2), In this code I have taken x(2) value as constant 0.5. Here da is a file having 0 and 1 (100 values; defined as damage state) and the other WH is file having values between 1.25-2.0 (100 values). Basically this value is computed using some formula 0 means no failure and 1 mean failure.

请先登录，再回答此问题。

采纳的回答

Torsten
Torsten 2022-5-2
编辑：Torsten 2022-5-2
da = load("C:\Users\Admin\OneDrive - IIT Bombay\Desktop\standard deviation\fragility1.txt")%%Damage state in terms of 0 and 1
WH = load("C:\Users\Admin\OneDrive - IIT Bombay\Desktop\standard deviation\velocity1.txt")%%
n = length(da);
xdamage = da(:,1);
wh = WH(:,1);
x0 = [2 0.5];%x valueis a median values%
%options = optimset('LargeScale','off','Display','off','TolX',0.001,'TolFun',0.001)
%[x,fval] = fminsearch(@myfun,x0,options,n,wh,xdamage)%fval is the likeli hood function%
[x,fval] = fminsearch(@(x)myfun(x,n,wh,xdamage),x0);%fval is the likeli hood function%
function f = myfun(x,n,wh,xdamage);
p1=0.0;
for i=1:n
yx=(log(wh(i)/x(1)))/x(2);%%x(2) is standard deviation
if yx >= 5.0;
y1=5.0;%%maximum value normcdf can take taken as 5%%
elseif yx<=-5.0;
y1=-5.0;
else
y1 = yx;
end
y2=normcdf(y1);
p1=p1+log(((y2)^xdamage(i))*((1.0-y2)^(1.0-xdamage(i))));%%Maximum likeli hood%
end
f=-p1;
end
  2 个评论
Raj Arora
Raj Arora 2022-5-2
Thanks torsten for your prompt reply. But this is still not giving appropriate answer the standard deviation and maximum likelihood value coming are very small which doesnot seems correct to me.
Torsten
Torsten 2022-5-2
编辑：Torsten 2022-5-2
I didn't check your equations for correctness - I only gave you the way how to work with two inputs for fminseach.
If you describe in more detail what you are doing in your code, maybe someone can also help in terms of content.

请先登录，再进行评论。

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Multiobjective Optimization 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