MA 161 - Lesson 28: Linear Approximation and Differentials (4.6)

This lesson covers linear approximation of functions using tangent lines to approximate function values, as well as error and percentage error calculations.

Warm-up: Finding Equation of a Tangent Line

Warm-up Problem

Find the equation of the tangent line to \(f(x) = \sqrt{x}\) at the point \((4, 2)\).

Graph Description: The coordinate plane shows the function \(f(x) = \sqrt{x}\) graphed as a blue curve starting at the origin and increasing with a decreasing slope. At the point \((4, 2)\), marked with a dot, a red tangent line touches the curve. The tangent line has a positive slope and extends beyond the point of tangency in both directions. Dashed lines extend from the point \((4, 2)\) to both axes, marking the coordinates clearly.

Solution:

To find the equation of the tangent line, we need the point \((4, 2)\) and the slope at that point.

First, find the derivative: \(f'(x) = \frac{1}{2\sqrt{x}}\)

Evaluate at \(x = 4\): \(f'(4) = \frac{1}{2\sqrt{4}} = \frac{1}{4}\)

Using point-slope form: \(y - 2 = \frac{1}{4}(x - 4)\)

Simplifying: \(y = \frac{1}{4}x + 1\)

Linear Approximation

Definition: Linear Approximation

The linear approximation (or tangent line approximation) of a function \(f(x)\) at a point \(x = a\) is given by:

\[L(x) = f(a) + f'(a)(x - a)\]

This equation represents the tangent line to \(f(x)\) at the point \((a, f(a))\), and can be used to approximate values of \(f(x)\) for \(x\) near \(a\).

Key Concept: Linear approximation works best when \(x\) is close to \(a\). The farther \(x\) is from \(a\), the less accurate the approximation becomes.

Examples of Linear Approximation

Example 1: Linear Approximation of Sine Function

Find the linear approximation of \(f(x) = \sin(x)\) at \(x = 0\).

Solution:

We need \(f(0)\) and \(f'(0)\).

Since \(f(x) = \sin(x)\), we have \(f(0) = \sin(0) = 0\)

The derivative is \(f'(x) = \cos(x)\), so \(f'(0) = \cos(0) = 1\)

The linear approximation is:

\[L(x) = f(0) + f'(0)(x - 0) = 0 + 1 \cdot x = x\]

Therefore, \(\sin(x) \approx x\) for \(x\) near 0.

Example 2: Using Linear Approximation to Estimate Values

Use the linear approximation of \(f(x) = \sin(x)\) at \(x = 0\) to approximate the values of \(\sin(\pi/6)\), \(\sin(\pi/3)\), and \(\sin(\pi/12)\).

Solution:

From Example 1, we have \(L(x) = x\), so \(\sin(x) \approx x\) for \(x\) near 0.

For \(\sin(\pi/6)\): \(L(\pi/6) = \pi/6 \approx 0.524\)

(Actual value: \(\sin(\pi/6) = 0.5\))

For \(\sin(\pi/3)\): \(L(\pi/3) = \pi/3 \approx 1.047\)

(Actual value: \(\sin(\pi/3) \approx 0.866\))

For \(\sin(\pi/12)\): \(L(\pi/12) = \pi/12 \approx 0.262\)

(Actual value: \(\sin(\pi/12) \approx 0.259\))

Observation: The approximation is most accurate for \(\sin(\pi/12)\) because \(\pi/12\) is closest to 0. The approximation becomes less accurate as we move farther from the point of tangency.

Example 3: Estimating Square Roots

Use linear approximation to estimate the value of \(\sqrt{40}\).

Solution:

Let \(f(x) = \sqrt{x}\). We need to choose a point \(a\) near 40 where we can easily calculate \(f(a)\).

Choose \(a = 36\) (since \(\sqrt{36} = 6\) is easy to compute).

Find the derivative: \(f'(x) = \frac{1}{2\sqrt{x}}\)

Evaluate at \(a = 36\): \(f'(36) = \frac{1}{2\sqrt{36}} = \frac{1}{12}\)

The linear approximation is:

\[L(x) = f(36) + f'(36)(x - 36) = 6 + \frac{1}{12}(x - 36)\]

To estimate \(\sqrt{40}\), substitute \(x = 40\):

\[L(40) = 6 + \frac{1}{12}(40 - 36) = 6 + \frac{4}{12} = 6 + \frac{1}{3} \approx 6.333\]

Therefore, \(\sqrt{40} \approx 6.333\)

(Actual value: \(\sqrt{40} \approx 6.325\))

Example 4: Estimating Natural Logarithms

Use linear approximation to estimate the value of \(\ln(1.2)\).

Solution:

Let \(f(x) = \ln(x)\). Choose \(a = 1\) since \(\ln(1) = 0\) is easy to compute.

Find the derivative: \(f'(x) = \frac{1}{x}\)

Evaluate at \(a = 1\): \(f'(1) = \frac{1}{1} = 1\)

The linear approximation is:

\[L(x) = f(1) + f'(1)(x - 1) = 0 + 1(x - 1) = x - 1\]

To estimate \(\ln(1.2)\), substitute \(x = 1.2\):

\[L(1.2) = 1.2 - 1 = 0.2\]

Therefore, \(\ln(1.2) \approx 0.2\)

(Actual value: \(\ln(1.2) \approx 0.182\))

Example 5: Polynomial Function

Given \(f(x) = 8 - 3x^2\), use linear approximation at \(x = 2\) to estimate \(f(1.9)\).

Solution:

First, evaluate \(f(2)\): \(f(2) = 8 - 3(2)^2 = 8 - 12 = -4\)

Find the derivative: \(f'(x) = -6x\)

Evaluate at \(x = 2\): \(f'(2) = -6(2) = -12\)

The linear approximation is:

\[L(x) = f(2) + f'(2)(x - 2) = -4 + (-12)(x - 2) = -4 - 12(x - 2)\]

To estimate \(f(1.9)\), substitute \(x = 1.9\):

\[L(1.9) = -4 - 12(1.9 - 2) = -4 - 12(-0.1) = -4 + 1.2 = -2.8\]

Therefore, \(f(1.9) \approx -2.8\)

(Actual value: \(f(1.9) = 8 - 3(1.9)^2 = 8 - 10.83 = -2.83\))

Example 6: Comparing Approximations

Estimate the value of \(\cos(11\pi/8)\) using linear approximation of \(f(x) = \cos(x)\) at both \(x = \pi/4\) and \(x = \pi/2\). Which approximation is better?

Solution:

Approximation at \(x = \pi/4\):

We have \(f(\pi/4) = \cos(\pi/4) = \frac{\sqrt{2}}{2}\)

The derivative is \(f'(x) = -\sin(x)\), so \(f'(\pi/4) = -\sin(\pi/4) = -\frac{\sqrt{2}}{2}\)

Linear approximation:

\[L_1(x) = \frac{\sqrt{2}}{2} - \frac{\sqrt{2}}{2}(x - \pi/4)\]

Substituting \(x = 11\pi/8\):

\[L_1(11\pi/8) = \frac{\sqrt{2}}{2} - \frac{\sqrt{2}}{2}(11\pi/8 - \pi/4)\] \[= \frac{\sqrt{2}}{2} - \frac{\sqrt{2}}{2}(11\pi/8 - 2\pi/8)\] \[= \frac{\sqrt{2}}{2} - \frac{\sqrt{2}}{2} \cdot \frac{9\pi}{8}\]
Approximation at \(x = \pi/2\):

We have \(f(\pi/2) = \cos(\pi/2) = 0\)

\(f'(\pi/2) = -\sin(\pi/2) = -1\)

Linear approximation:

\[L_2(x) = 0 + (-1)(x - \pi/2) = -(x - \pi/2)\]

Substituting \(x = 11\pi/8\):

\[L_2(11\pi/8) = -(11\pi/8 - \pi/2) = -(11\pi/8 - 4\pi/8) = -\frac{7\pi}{8}\]

Comparison: Note that \(11\pi/8 \approx 4.32\), \(\pi/4 \approx 0.785\), and \(\pi/2 \approx 1.571\). Since \(11\pi/8\) is much closer to \(\pi/2\) than to \(\pi/4\), the approximation at \(x = \pi/2\) should be more accurate. However, in practical terms, \(11\pi/8\) is still relatively far from both points, so neither approximation will be highly accurate.

Key Principles for Linear Approximation

Steps for Linear Approximation:
  1. Identify the function \(f(x)\) and the value you want to approximate.
  2. Choose a point \(a\) close to your target value where \(f(a)\) is easy to compute.
  3. Find the derivative \(f'(x)\) and evaluate \(f'(a)\).
  4. Write the linear approximation: \(L(x) = f(a) + f'(a)(x - a)\).
  5. Substitute your target value into \(L(x)\) to get the approximation.

Accuracy of Approximations: The accuracy of a linear approximation depends on:

Error and Percentage Error

Definition: Absolute Error

The absolute error in a linear approximation is:

\[\text{Error} = |f(x) - L(x)|\]

where \(f(x)\) is the actual value and \(L(x)\) is the approximated value.

Definition: Percentage Error

The percentage error is:

\[\text{Percentage Error} = \frac{|f(x) - L(x)|}{|f(x)|} \times 100\%\]

This expresses the error as a percentage of the actual value.

Common Mistake: Don't confuse the error in the approximation with the value \(x - a\). The error \(|f(x) - L(x)|\) measures how far the approximation is from the true value, while \(x - a\) measures the distance from the point of tangency.

Summary

Key Takeaways: