How to use Newton’s method to approximate a root
If we have a starting guess for a root, a tangent line can give us a new guess. We’ll see where that next guess comes from, then use it to work through four steps.
Use the tangent line to get the next estimate
A root is an -value where . Start at an estimate . The tangent line at has slope . Set the tangent line’s height equal to zero and solve for its -intercept to get .
Use this step where the derivative exists and is nonzero. The starting guess matters: the method can go to a different root or fail to converge. A graph can help you choose a starting point near the root you want.
Start at −1 and find the first approximation
For , move all terms to one side. Use and . At , the function value is and the derivative is .
Use each new estimate in the next step
Use the exact value to find . At , the function value is and the derivative is . Substitute both into the same formula.
For the next two steps, it helps to simplify the update. Put the two terms over the common denominator .
Substitute to get , then substitute to get . The table keeps the exact fractions alongside the rounded values.
| Step | Exact estimate | Four decimals |
|---|---|---|
| 1 | −0.3333 | |
| 2 | −0.0667 | |
| 3 | −0.0039 | |
| 4 | 0.0000 |
The fourth estimate rounds to . Here we can also check directly that . The equation has another root at , and our starting value of led us toward the root at .