cusum算法 matlab,求解二维马尔科夫链

先确认几件事

1. elseif k~=0&&l~=0

temp=min(b2,max(a1,b1));

T1=normcdf(temp,0,1);

temp=max(b1,max(a2,b2));  OR temp=max(b1,min(a2,b2));

2.a1=(k-i-0.5)*w*(log(1+250*(m+l*q)^2+2.332.....

a2=(k-i+0.5)*w*(log(1+250*(m+l*q)^2+2.332.....

250如何定的

这会影响结果

3.我先把probability matrix

放进,你可试试

m1=17;

m2=25;

R=zeros(m1*m2,m1*m2);

for i=0:1:m1-1

for j=0:1:m2-1

for k=0:1:m1-1

for l=0:1:m2-1

a1=(k-i-0.5)*w*(log(1+250*(m+l*q)^2+2.332*(m+l*q)/2)/(m+l*q)-1.166)+(m+l*q)/2;

a2=(k-i+0.5)*w*(log(1+250*(m+l*q)^2+2.332*(m+l*q)/2)/(m+l*q)-1.166)+(m+l*q)/2;

b1=(m+(l-0.5)*q-(1-p)*(m+j*q))/p;

b2=(m+(l+0.5)*q-(1-p)*(m+j*q))/p;

if k==0&&l==0

temp=min(a2,b2);

T=normcdf(temp,0,1);

R(i*m2+(j+1),k*m2+(l+1))=

elseif k==0&&l~=0

temp=b1;

T1=normcdf(temp,0,1);

temp=max(b1,min(a2,b2));

T2=normcdf(temp,0,1);

R(i*m2+(j+1),k*m2+(l+1))=

elseif k~=0&&l==0

temp=a1;

T1=normcdf(temp,0,1);

temp=max(a1,min(a2,b2));

T2=normcdf(temp,0,1);

R(i*m2+(j+1),k*m2+(l+1))=

elseif k~=0&&l~=0

temp=min(b2,max(a1,b1));

T1=normcdf(temp,0,1);

temp=max(b1,min(a2,b2));

T2=normcdf(temp,0,1);

R(i*m2+(j+1),k*m2+(l+1))=

end;  %end if

end;

end;

end;

end;