Calculus Methods

13 Newton's Method

 Back to Dr. Nandor's Calculus Methods Page

 Back to Dr. Nandor's Calculus Page

 

 

                This is a method for finding the zeroes of a function. Is it

                useful? Not so much. There are much better methods out

                there at this time in history, and any graphing calculator

                will do it as well. Why include it, then? Well, it's on the

                AP Calculus BC test, so here it is. For a brief, probably

                unhelpful derivation of why it works, see lesson 19.

 

 

        1) A brief sketch might help, but for a complicated function,

        this would be time-prohibitive.

 

        2) Find the first derivative of the function.

 

        3) Pick a starting value for x that is anywhere near the actual

        zero. The guess doesn't have to be great, but it should be as

        good as possible. Call this starting guess .

 

        4) The next step is to find using .

 

        5) Continue along in the same way, .

 

        6) The process continues until some tolerance is reached

        ( is less than some value).

 

        7) Repeat for as many zeroes as you need to find!

 

 

        Example: Find the roots of to within 0.05. Note

        that since this is a simple function, we already know that the

        answers should be .

 

        1) Let's skip the sketch.

 

        2)

 

        3) Let's start with .

 

        4)

 

 

            5)

 

 

            

 

 

        6) Note that , so we are finished with one

        of the zeroes (). At this point, however, we would

        probably look at the answer and say to ourselves "hey, that

        looks a lot like -1, let's try plugging -1 in to the equation to see

        if the answer is zero. So using this information, we would

        discover that one of the zeroes occurs at .

 

        7) Now we repeat for another zero. Since we are dealing with

        a quadratic, we could have two zeroes.

 

 

        1) We'll still skip the drawing.

 

        2)

 

        3) .

 

        4)

 

        5)

 

        

 

            

 

 

        6) Again, we see that and within our

        tolerance.

 

 

        So our final answer is that the two zeroes that we found using

        Newton's Method are and , but we would see

        how close these are to whole numbers, and after checking

        those numbers in the original equation, we would know that

        the two answers are and .

 

 

On to Method 14 - Simple Antiderivatives

 Back to Dr. Nandor's Calculus Methods Page

 Back to Dr. Nandor's Calculus Page