Math help

[Darn]

I don't quite understand my MATLAB homework here. How do I interpret the provided differential equation for the epidemiology, to plot the functions? For example, from R'(t? I think?)=nI-uR I would get the antiderivative R(t)=nIt-uRt. Or something. Which seems... odd. Really, I'm at a loss here. I know there are plenty of Superior Math People here, so... Help? What exactly am I supposed to do here?

[Feil]

I and R are functions, not constants. Functions do not integrate like constants. Your given equation, R'=nI - uR is a first order linear differential equation. Solve it like one. There is a good tutorial here: http://www.sosmath.com/diffeq/first/lin ... eareq.html

[Darn]

Thanks a lot. This is a little embarrassing, I really ought to have recognized basic differential equations... But that class was sooo long ago... Anyway, when I knew what I was looking at, I actually managed to figure out how to make matlab do the heavy lifting for me (with ode23), but now the damn lambda is causing me trouble. I mean, if I read the damn exercise right, lambda=max(200*I*I,10^-6), but plugging that into the code makes ode23 cry... I'm not sure if the result is even supposed to be solvable. Help, again?

[Feil]

max(A, B) tells the computer to look at A, look at B, and output the bigger value. It can't do it if A and B aren't matrices of the same dimensions, e.g. since your B is a scalar A must also be a scalar. Make sure you haven't defined I as a multi-valued matrix, like a vector or something. If I is a vector, you'll need to reduce it to its scalar length before you can compare it to a scalar.

Obviously the computer needs to know what I is before it can tell you whether 200I2 is bigger than 10-6. So you mustn't call lambda until after the computer has loaded I with a numerical value.

EDIT: If that's not an option, I guess you could make the computer run the model twice, once for L=200I2 and once for L=10-6, use those to load I with a number, then have the computer check which L to use after the fact, discard the incorrect run, and output the result of the correct run. There might be a more elegant way of doing this, but I don't know MATLAB.

[Kuroneko]

If you're doing arithmetic elementwise rather than as matrices, use the . prefix for the operator.
For example, l*l is l by l multiplied as matrices, while l.*l is the matrix consisting of each individual element multiplied by itself.
Similarly, if you're not too concerned about optimization, you can get the max of matrix A and scalar B as max(A,B*ones(size(A)).

[Darn]

I was a bit unclear, sorry. The "I" in question is one of the functions . As I said, I'm not sure if I'm reading the assignment correctly.