Linear Approximation

What Is Linear Approximation?

Linear approximation is a method for estimating the value of a function at a specific point by using the equation of a line that is tangent to the function at a nearby, "easier" point. In simpler terms, we find a line that closely hugs a curve at a certain spot and use that simple line to guess the curve's values nearby.

Think about looking at a globe from space. It's clearly a sphere. But if you zoom in on a map of your neighborhood, the ground looks flat. On that small scale, a flat surface (like a line or a plane) is a great approximation of the Earth's curve. Linear approximation does the exact same thing with mathematical functions. We zoom in on a point on a graph until the curve looks like a straight line, and then we use that line to make calculations.

Why Do We Need Linear Approximation?

You might be thinking, "Why not just use a calculator?" That's a great question! While we have powerful tools today, linear approximation is still a vital concept for several reasons:

• Historical Importance: Before electronic calculators, mathematicians, scientists, and engineers relied on methods like linear approximation to calculate values for things like planetary orbits, bridge designs, and complex equations. It was a fundamental tool for scientific progress.
• Computational Speed: In computer graphics, video games, and scientific simulations, performing millions of calculations per second is necessary. Sometimes, calculating an exact value is too slow. Using a fast linear approximation can give a result that is visually or functionally identical, saving precious processing power.
• Understanding Change: Linear approximation is built on the idea of a function's slope, or rate of change, at a single point. This is a core concept in calculus and physics. Learning about it now builds a strong foundation for understanding how things change, from the speed of a car to the growth of an investment.
• Problem Solving: It provides a powerful way to simplify complex problems. In physics and engineering, many complicated formulas can be simplified into linear ones when you're only looking at small changes, making them much easier to solve.

The Key Idea: What Is a Tangent Line?

The hero of linear approximation is the tangent line. Imagine a winding roller coaster track representing the graph of our function, say, $f(x)=x^2$. A tangent line is like a straight plank of wood that you rest against the track at a single, specific point. At that exact point, the plank has the same steepness, or slope, as the track. It "just touches" the curve without crossing it (at least, not right there).

The tangent line at a point $(a,f(a))$ is the unique straight line that best represents the direction of the curve at that precise location. Because the line and the curve are so similar right near that point, the line's y-values will be very close to the curve's y-values. This is the secret that makes our approximation work.

How Do You Find the Equation for the Approximation?

To build our approximation, we need the equation of the tangent line. We can find this using the familiar point-slope form of a line: $y-y_1=m(x-x_1)$.

Let's adapt this for our purpose:

1. The Point $(x_1,y_1)$: This is our "nice" point of tangency, which we'll call $(a,f(a))$. This is a point where we can easily calculate the function's value.
2. The Slope $m$: This is the slope of the curve at $x=a$. This is the most important new idea. In calculus, this slope is called the derivative and is written as $f^{\prime }(a)$ (read as "f-prime of a"). For now, you can think of $f^{\prime }(x)$ as a special "slope formula" that we are given for each function. For example, for $f(x)=x^2$, the slope formula is $f^{\prime }(x)=2x$.

Substituting our parts into the point-slope form, we get:
$y-f(a)=f^{\prime }(a)(x-a)$

To get our final approximation formula, we just solve for $y$. We call this approximation function $L(x)$ for "Linearization":

This formula tells us that our approximation $L(x)$ starts at the known value $f(a)$ and then adds or subtracts a small amount based on the slope $f^{\prime }(a)$ and how far $x$ is from $a$, which is $(x-a)$.

Putting It All Together: Worked Examples

Let's use the formula to approximate some tricky values. For each problem, we will be given the function $f(x)$ and its slope formula $f^{\prime }(x)$.

How Good Is the Approximation?

The accuracy of a linear approximation depends entirely on one thing: how close $x$ is to $a$. The approximation is best right near the point of tangency where the line and curve are almost identical. As you move farther away from $a$, the curve starts to bend away from the tangent line, and the error in your estimation grows.

Let's look at our first example, $f(x)=\sqrt{x}$, and the tangent line at $a=9$. See how the error increases as we move away from 9.

As you can see, the error is tiny for values extremely close to 9, but it becomes noticeably larger by the time we get to 11. The takeaway is simple: for the best results, always choose a point $a$ that is as close as possible to the $x$ value you want to estimate.

Common Mistakes to Avoid

Linear approximation is a straightforward process, but a few common errors can trip you up. Be on the lookout for these:

• Choosing a Poor Point: If you need to estimate $\sqrt{99}$, choosing $a=1$ is a bad idea. The best choice is $a=100$ because it's very close and easy to calculate. Always pick the nearest, simplest point.
• Mixing up $x$ and $a$: The slope $f^{\prime }(a)$ must be calculated at your nice point $a$. A common mistake is to plug $x$ into the slope formula. Remember, the slope of your approximating line is constant; it's fixed at the point of tangency.
• Sign Errors: Be careful with the $(x-a)$ term. If $x$ is less than $a$ (like in our example with $x=3.9$ and $a=4$), this term will be negative. A simple sign error can throw off the entire result.
• Forgetting the $f(a)$ Term: The formula is $f(a)+f^{\prime }(a)(x-a)$. Don't forget to add your starting value! The approximation builds from the height of the function at point $a$.

Quick Summary: The 4-Step Process

To find a linear approximation, just follow these four steps:

1. Identify: Determine your function $f(x)$ and the target value $x$ you need to estimate.
2. Choose: Select a "nice" point $a$ very close to $x$ where you can easily find the value of $f(a)$.
3. Calculate: Find the two key pieces of information: the function's value $f(a)$ and the function's slope $f^{\prime }(a)$ at that nice point.
4. Approximate: Plug $f(a)$, $f^{\prime }(a)$, and $a$ into the linear approximation formula $L(x)=f(a)+f^{\prime }(a)(x-a)$ and then substitute your target value of $x$ to find the answer.