2.4. Secant Method

The Secant method is a variation of the Newton-Raphson method. Instead of using the derivative f′(x), the method approximates the derivative using two previous function values.

The Secant method offers an alternative approach to finding roots of a function. Unlike the Newton-Raphson method, which requires the exact derivative of the function, the Secant method estimates the derivative based on two previous function values. This makes it useful in situations where the derivative is difficult or impossible to calculate.

Imagine trying to find the height of a mountain. Instead of measuring the exact slope at a point, you take two measurements at different heights and use them to estimate how steep the mountain is. This is akin to how the Secant method uses two function values to approximate the derivative.

1. Start with two initial guesses x0 and x1.
2. Use the following iterative formula to compute the next approximation: xn+1=xn−f(xn)(xn−xn−1)f(xn)−f(xn−1)
3. Repeat the process until the difference between successive approximations is less than a desired tolerance: ∣xn+1−xn∣<ϵ.

To apply the Secant method, you begin with two guesses for the root, labeled x0 and x1. Using these points, you apply the provided iterative formula to calculate the next approximation (xn+1). This process repeats, refining the approximation each time. You continue iterating until the change between the new guess and the previous one is smaller than your set tolerance level, indicating you've found a sufficiently accurate root.

Consider this method like refining your guess on the depth of a pool. Suppose your first guess is based on two measurements or estimates from the pool’s edge. You use these to approximate the depth and keep adjusting your guess until you are confident it's accurate enough.

● Advantages: ○ Does not require the computation of the derivative. ○ Can converge faster than the Bisection method, though slower than Newton-Raphson. ● Disadvantages: ○ Requires two initial guesses. ○ May fail to converge if the two initial guesses are not appropriate.

One significant advantage of the Secant method is that it circumvents the need for calculating derivatives, easing its application in practical scenarios. It may provide faster convergence compared to the slower Bisection method, though it typically does not reach the speed of convergence associated with the Newton-Raphson method. However, it does have its drawbacks; specifically, it relies on two initial guesses, which means what you start with matters greatly. If those guesses are poor, the method may not converge to the root at all.

Imagine you're navigating to a new city using two landmarks (your initial guesses). If those landmarks are well-placed, you can find your way quickly. However, if they're too far apart or incorrectly chosen, you might end up wandering, unable to reach your destination. Similarly, your choices in the Secant method significantly influence your success.

For f(x)=x2−4: ● Initial guesses: x0=1, x1=3. ● Using the formula: x2=3−f(3)(3−1)f(3)−f(1)=3−(9−4)(2)(9−4)−(1−4)=3−106≈1.8333. ● Repeat the process until convergence to x=2.

Let's work through the Secant method using the function f(x) = x^2 - 4. First, we set our initial guesses: x0=1 and x1=3. We then calculate the next approximation, x2, by applying the iterative formula: it uses both function values (f(3) and f(1)) to compute. You repeat similar computations, refining your estimates until you zero in on the root at x=2.

Think of it like adjusting the position of a sliding door. You start with two positions of the door—where you think is open (x0) and where you know it’s closed (x1). Each time you try, you adjust your estimate of where the door truly opens further, progressively getting closer to that exact open point (the root).

--