

Similarly we can construct a function, either inline or anonymous, from the derivative of f: fxin=inline(char(diff(f))) We can create such a function with fanon = subs(f, x, t) In MATLAB 7, the best way to make f into function is with the operator, creating an anonymous function. (If we had simply typed fin=inline(f) we'd get an error message, since f is not a string.) The inline function fin now accepts an argument: fin(4) What's going on here is that the inline command requires a string as an input, and char turns f from a symbolic expression to the string 'x^2-sin(x)'.

We can also turn f into an inline function with the command: fin=inline(char(f)) We can evaluate f(4) by substituting 4 for x using the subs command: subs(f,x,4)

? Error using => mupadmex Error in MuPAD command: Index exceeds matrix dimensions.Įrror in => sym.sym>sym.subsref at 1366 B = mupadmex('mllib::subsref',A.s,inds) However, we cannot evaluate it, at least in the obvious way. In the case of several symbolic variables, we can specify the one with respect to which we want to differentiate. MATLAB recognizes what the "variable" is. The next line shows we can differentiate f in the obvious way.

That is, if we know that f(x) = x^2, we know that f(4) = 4^2 = 16. On the other hand, any symbolic expression implies a rule for evaluation. This definition is more general for example, it allows us to define a function f(x) to be x^2 if x is negative or 0, and sin(x) if x is positive. The other is a rule (algorithm) for producing a numerical output from a given numerical input or set of numerical inputs. One is a symbolic expression such as sin(x) or x^2. There are two distinct but related notions of function that are important in Calculus. In this published M-file, we show how to represent functions of one varaible in MATLAB,then how to do symbolic computation, numerical computation and plotting with them.
