Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is 1-by-15. - can't find what the problem is

I keep getting this error, but I have no idea what's causing it.

Using the step button brings up nothing until the very end when it tries to finish.

all of the X's,Z's,Y's within Fd are 15x1 with positive integers

Any ideas as to what's causing this?

Fd=zeros(cap, cap, cap,term); %here is the original matrix I created       cap=15 and term=20
...
    
for tt=19:-1:1
    for i=1:cap
    for j=1:cap
        for k=1:cap
            
                ... %for brevity's sake I've ommitted some stuff here that doesn't seem to be causing issues

        Fd(i,j,k,tt)= b*(... 
            m1*(...
                p(y(j),z1(k))*(n1*Fd(xp(i),zp(j),y(k),tt+1) + (1-n1)*Fd(xpp(i),zpp(j),y(k),tt+1))+...
                (1-p(y(j),z1(k)))*(n2*Fd(xp(i),z1(j),yg(k),tt+1) + (1-n2)*Fd(xpp(i),z1(j),ygg(k),tt+1)))+...
                (1-m1)*(... 
                     p2(y(j),z2(k))*(n1*Fd(xp(i),zd(j),y(k),tt+1) + (1-n1)*Fd(xpp(i),zdd(j),y(k),tt+1))+...
                (1-p2(y(j),z2(k)))*(n2*Fd(xp(i),z2(j),yg(k),tt+1) + (1-n2)*Fd(xpp(i),z2(j),ygg(k),tt+1))...
                    ))+...
            (1-b)*( ...
                m2*(...
                p(y1(j),z(k))*(n1*Fd(xp(i),zg(j),y1(k),tt+1) + (1-n1)*Fd(xpp(i) ,zgg(j),y1(k),tt+1))+ ...
                (1-p(y1(j),z(k)))*(n2*Fd(xp(i),z(j),yp(k),tt+1) + (1-n2)*Fd(xpp(i) ,z(j),ypp(k),tt+1)))+ ...
                (1-m2)*(p(y2(j),z(k))*(n1*Fd(xp(i),zg(j),y2(k),tt+1) + (1-n1)*Fd(xpp(i),zgg(j),y2(k),tt+1))+...
            (1-p(y2(j),z(k)))*(n2*Fd(xp(i),z(j),yd(k),tt+1) + (1-n2)*Fd(xpp(i),z(j),ydd(k),tt+1))));

If more code is required, I can certainly provide

[Torsten]

If you don't know which variable on the right-hand side is 15x1, you will have to output their sizes within the loop.

But I'm almost sure that having Fd on both sides of the equation will cause problems.

[Kitt]

So you think it's because I have Fd on both sides? I'm brand spanking new to matlab and I'm trying to run a stochastic dynamic program. It's been a steep learning curve.

Answer 1

[Voss]

Assuming that b, m1, m2, n1, and n2 are scalars, the code runs ok. (Also assuming that the integers stored in the x, y, z variables do not exceed 15; if they did you'd get a different error.) If any of b, m1, m2, n1, or n2 are not scalars, you might get the error you saw.

Running with scalars works:

cap = 15;
term = 20;

b = 0.8;
m1 = 0.7;
m2 = 0.6;
n1 = 0.5;
n2 = 0.4;

xp = randi(cap,15,1);
xpp = randi(cap,15,1);

y = randi(cap,15,1);
y1 = randi(cap,15,1);
y2 = randi(cap,15,1);
yd = randi(cap,15,1);
ydd = randi(cap,15,1);
yg = randi(cap,15,1);
ygg = randi(cap,15,1);
yp = randi(cap,15,1);
ypp = randi(cap,15,1);

z = randi(cap,15,1);
z1 = randi(cap,15,1);
z2 = randi(cap,15,1);
zd = randi(cap,15,1);
zdd = randi(cap,15,1);
zg = randi(cap,15,1);
zgg = randi(cap,15,1);
zp = randi(cap,15,1);
zpp = randi(cap,15,1);

p = rand(cap,cap);
p2 = rand(cap,cap);

Fd=zeros(cap, cap, cap, term);

for tt=term-1:-1:1
    for i=1:cap
        for j=1:cap
            for k=1:cap
                
                Fd(i,j,k,tt) = ...
                    b*( ... 
                        m1*( ...
                            p(y(j),z1(k))*(n1*Fd(xp(i),zp(j),y(k),tt+1) + (1-n1)*Fd(xpp(i),zpp(j),y(k),tt+1))+ ...
                            (1-p(y(j),z1(k)))*(n2*Fd(xp(i),z1(j),yg(k),tt+1) + (1-n2)*Fd(xpp(i),z1(j),ygg(k),tt+1)))+ ...
                        (1-m1)*( ...
                            p2(y(j),z2(k))*(n1*Fd(xp(i),zd(j),y(k),tt+1) + (1-n1)*Fd(xpp(i),zdd(j),y(k),tt+1))+ ...
                            (1-p2(y(j),z2(k)))*(n2*Fd(xp(i),z2(j),yg(k),tt+1) + (1-n2)*Fd(xpp(i),z2(j),ygg(k),tt+1)) ...
                            ))+ ...
                    (1-b)*( ...
                        m2*( ...
                            p(y1(j),z(k))*(n1*Fd(xp(i),zg(j),y1(k),tt+1) + (1-n1)*Fd(xpp(i) ,zgg(j),y1(k),tt+1))+ ...
                            (1-p(y1(j),z(k)))*(n2*Fd(xp(i),z(j),yp(k),tt+1) + (1-n2)*Fd(xpp(i) ,z(j),ypp(k),tt+1)))+ ...
                        (1-m2)*( ...
                            p(y2(j),z(k))*(n1*Fd(xp(i),zg(j),y2(k),tt+1) + (1-n1)*Fd(xpp(i),zgg(j),y2(k),tt+1))+ ...
                            (1-p(y2(j),z(k)))*(n2*Fd(xp(i),z(j),yd(k),tt+1) + (1-n2)*Fd(xpp(i),z(j),ydd(k),tt+1)) ...
                            ));
                
            end
        end
    end
end

any(Fd,'all')

ans = logical
   0

Running with b, for instance, of size 15x1 generates the error you got:

b = 0.8*ones(cap,1);

Fd=zeros(cap, cap, cap, term);

for tt=term-1:-1:1
    for i=1:cap
        for j=1:cap
            for k=1:cap
                
                Fd(i,j,k,tt) = ...
                    b*( ... 
                        m1*( ...
                            p(y(j),z1(k))*(n1*Fd(xp(i),zp(j),y(k),tt+1) + (1-n1)*Fd(xpp(i),zpp(j),y(k),tt+1))+ ...
                            (1-p(y(j),z1(k)))*(n2*Fd(xp(i),z1(j),yg(k),tt+1) + (1-n2)*Fd(xpp(i),z1(j),ygg(k),tt+1)))+ ...
                        (1-m1)*( ...
                            p2(y(j),z2(k))*(n1*Fd(xp(i),zd(j),y(k),tt+1) + (1-n1)*Fd(xpp(i),zdd(j),y(k),tt+1))+ ...
                            (1-p2(y(j),z2(k)))*(n2*Fd(xp(i),z2(j),yg(k),tt+1) + (1-n2)*Fd(xpp(i),z2(j),ygg(k),tt+1)) ...
                            ))+ ...
                    (1-b)*( ...
                        m2*( ...
                            p(y1(j),z(k))*(n1*Fd(xp(i),zg(j),y1(k),tt+1) + (1-n1)*Fd(xpp(i) ,zgg(j),y1(k),tt+1))+ ...
                            (1-p(y1(j),z(k)))*(n2*Fd(xp(i),z(j),yp(k),tt+1) + (1-n2)*Fd(xpp(i) ,z(j),ypp(k),tt+1)))+ ...
                        (1-m2)*( ...
                            p(y2(j),z(k))*(n1*Fd(xp(i),zg(j),y2(k),tt+1) + (1-n1)*Fd(xpp(i),zgg(j),y2(k),tt+1))+ ...
                            (1-p(y2(j),z(k)))*(n2*Fd(xp(i),z(j),yd(k),tt+1) + (1-n2)*Fd(xpp(i),z(j),ydd(k),tt+1)) ...
                            ));
                
            end
        end
    end
end

Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is 15-by-1.

It may be that one or more of b, m1, m2, n1, and n2 has its value changed inadvertently to a non-scalar value.

You can try

dbstop if error

and when the code stops check the variables to figure out what's causing the problem.

[Torsten]

Try running the code without preallocating Fd :-)

[Voss]

OK, that's going to be a problem, but OP preallocates it in his code.

[Torsten]

Yes, but all these equations are implicit in the Fd values. Inserting 0 when the corresponding value for Fd is not yet determined doesn't make sense. All these equations in the loop had to be arranged to one big linear system with the Fd's as unknowns.

[Voss]

I'm sure you're right. I was just trying to reproduce the error.