Secant Method
Secant Method
The secant method also known as the 2-point method in numerical analysis is a
root-finding algorithm that uses a succession of roots of secant lines to better approximate a root
of a function f. The Secant Method is a practical and efficient way to approximate roots of
functions, particularly when the function's derivative is not readily available or when calculating
the derivative is computationally expensive. However, it's essential to choose initial guesses
carefully to ensure convergence, as the method may not always converge if the initial points are
too far apart or if the function exhibits particular behavior near the root.
Since the equation given has a positive ax^2 then we can determine that it is a parabola that
opens upwards from the point -3104 in the y-axis.
From the graph we can determine that the root should be approximately equal to 55.714.
𝑥1−𝑥0
Using the secant method: 𝑥2 = 𝑥1 − 𝑓(𝑥1) 𝑓(𝑥1)−𝑓(𝑥0)
Where we have the first initial guess x0 and the second initial guess x1, x2 shall be the
value which will converge to determine the root of the equation.
Algorithm of Secant Method
● Start with two initial approximations, x0 and x1, such that f(x0) and f(x1) have different
signs, i.e., f(x0) * f(x1) < 0. These initial points define an interval containing the root.
● For each iteration, compute the next approximation, x2, using the formula: x2 = x1 -
(f(x1) * (x1 - x0)) / (f(x1) - f(x0))
● Check for convergence. You can use various stopping conditions, such as a specified
number of iterations, a maximum allowed error, or a tolerance level
● If the stopping condition is met, consider x2 as the approximate root and stop the
algorithm.
● If the stopping condition is not met, update the values of x0 and x1 as follows: x0 = x1,
x1 = x2
● Go back to step 2 and repeat the process until the convergence criteria are satisfied.
2
Figure 2: Result for the equation x -3104 using the secant method
The result obtained from the secant method implementation is a root approximation of
the function f(x) = x^2 - 3104. Starting with the initial guesses of 1 and 2 and a tolerable error of
0.001, the secant method went through 12 iterations to narrow down the interval where the root
exists. The iterative process, as shown in the output, effectively divided the interval into smaller
subintervals, systematically approaching the root. The final result, which is approximately
55.71355307, represents a very close approximation to the actual root of the function. This result
demonstrates the secant method's reliability and accuracy in finding roots, provided the initial
interval brackets the root and the specified tolerance level is met. It's essential to choose initial
guesses and a tolerance level carefully to ensure the success of the method, and this result
demonstrates the effectiveness of the bisection method in finding the root of the function f(x) =
x^2 - 3104 and locating roots with precision. The initial guesses and the specified tolerable error
were used to perform a series of iterations, each step further narrowing down the interval where
the root is located. The method successfully converged to an approximation of the root, which is
approximately 55.71355307. What makes this method particularly advantageous is its
independence from the need for derivative information. Moreover, its versatility allows it to be
applied to a wide range of functions, even those that might not be differentiable but are
continuous. One of its strengths lies in its potential for rapid convergence, especially when
well-chosen initial approximations are employed. It often manages to find a root in just a few
iterations, which can result in computational efficiency. The Secant Method's simplicity in terms
of both understanding and implementation makes it accessible to a broad audience, enabling
individuals without advanced mathematical or programming skills to utilize it effectively.
References
Admin. (2022). Secant method - definition, procedure, and example. BYJUS.
https://byjus.com/maths/secant-method/