Function isn't returning anything

At the moment I have this function which I would like to determine if there is a critical point of the function f on the interval on the interval [a,b]. If there is I'd like it to return 1 for a relative max and -1 for a relative min, and 0 if it is not on the interval. At the moment I do not have the anything for returning the 0 because currently it only spits out v and ac given they are not suppressed for debugging. I can't really see any reason it isn't giving me something for n. Any and all help is much appreciated.

function n = myanalyzecp(f,a,b)

syms x i

v = diff(f(x))

ac = diff(diff(f(x)))

for i=[a:b]

if subs(v,a)==0

      if subs(ac,a)>0

          n==1

          return

      else if subs(ac,a)<0

          n==-1

          return

          end

      end

end 

end

Answer 1

[Geoff Hayes]

QingQangQong - note that in your code, you are using n as either

 n==1

or

 n==-1

The == is used when you want to check to see if two objects are identical. If you wish to assign 1 or -1 to n then use the equals sign

 n = 1;

or

 n = -1;

But even that might not fix the problem as you may not always satisfy either the if or the elseif conditions. In that case, you should assign a default value for n prior to entering the for loop.

[QingQangQong]

Thank you for your reply. I did as you said and changed the '==' to '=' and assigned n to a default value. Although I'm using a case which should return -1 (x^2,-3,3) it is still just returning the default value. I tried changing v to v(x) in the if statement and it threw an error. But the only thing I can think of that would be wrong as to why it isn't entering the if statement is the condition. Thoughts?

[Geoff Hayes]

Try stepping through the code (using the debugger). On each iteration of the for loop, check to see what

 subs(v,a)

is returning. Is this a double which you can compare with zero, or is it of some other data type?

Note that trying to check whether a floating point value (if that is what subs(v,a) is returning) is identical to zero is not appropriate/possible due to the nature of the data type (with integers you can do this comparison, but not with doubles). When comparing in this fashion, you should check to see if the value is close to zero (within some tolerance), and if so, then consider it to be zero. Something like

 tol = 0.000001;
 if abs(subs(v,a)) < tol
      if subs(ac,a)>0
          n==1
          return
      else if subs(ac,a)<0
          n==-1
          return
          end
      end
 end

tol is just the tolerance you use to determine whether the number can be considered zero or not.

[QingQangQong]

I think I got it. Silly mistake, I should have been substituting 'i' in not 'a'. Thank you for your help though!

Edit: Currently having an issue where the case (-2*x^2+6*x-2,1,7) returns the default n value. I presume this is because it has a max/min at a non-integer value. Is there anyways to make a for loop have non-integer step sizes.

Edit 2: Figured that out too. Just needed to add step size in between a and b.