Solved Exercises on Functions Represented as Power Series in Calculus 2

Master functions represented as power series: Taylor series, Maclaurin series, differentiation and integration of power series through these 5 detailed exercises.

Solution: Exercises 1 to 3
1 Geometric Series Representation
Exercise 1
Find a power series representation for:
\(f(x) = \frac{1}{1-x}\) and determine its interval of convergence.
Definition:

Power Series Representation: A function expressed as an infinite sum \(\sum_{n=0}^{\infty} c_n(x-a)^n\)

Geometric Series: \(\sum_{n=0}^{\infty} ar^n = \frac{a}{1-r}\) for \(|r| < 1\)

Power series construction method:
  1. Recognize the function as a geometric series form
  2. Identify the first term (a) and common ratio (r)
  3. Write the series expansion
  4. Determine convergence conditions
Original function
\(f(x) = \frac{1}{1-x}\)
Geometric form
\(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n\)
Step 1: Recognize geometric series pattern

The function \(\frac{1}{1-x}\) is in the standard form \(\frac{a}{1-r}\) where \(a = 1\) and \(r = x\)

Step 2: Apply geometric series formula

Using the geometric series formula: \(\frac{1}{1-r} = \sum_{n=0}^{\infty} r^n\) for \(|r| < 1\)

Substituting \(r = x\): \(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n\)

Step 3: Determine convergence condition

For convergence: \(|r| < 1 \Rightarrow |x| < 1\)

So the interval of convergence is \((-1, 1)\)

Step 4: Verify with partial sums

For small values of x, partial sums approximate the function well

\(S_N = 1 + x + x^2 + ... + x^N = \frac{1-x^{N+1}}{1-x}\)

As \(N \to \infty\), if \(|x| < 1\), then \(x^{N+1} \to 0\), so \(S_N \to \frac{1}{1-x}\)

Power series: \(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n = 1 + x + x^2 + x^3 + ...\)
Interval of convergence: \((-1, 1)\)
Final answer:

The power series representation is \(\sum_{n=0}^{\infty} x^n\) with interval of convergence \((-1, 1)\).

Applied rules:

• Geometric Series Formula: \(\frac{1}{1-r} = \sum_{n=0}^{\infty} r^n\) for \(|r| < 1\)

• Radius of Convergence: Determined by the condition \(|r| < 1\)

• Term-by-term: Each coefficient is 1 in this case

2 Integration of Power Series
Exercise 2
Find a power series representation for:
\(f(x) = \ln(1+x)\) by integrating the geometric series.
Definition:

Integration of Series: \(\int \sum_{n=0}^{\infty} c_n x^n dx = \sum_{n=0}^{\infty} \int c_n x^n dx = \sum_{n=0}^{\infty} \frac{c_n x^{n+1}}{n+1} + C\)

Term-by-term Integration: Valid within the radius of convergence

Start with
\(\frac{1}{1+x} = \sum_{n=0}^{\infty} (-1)^n x^n\)
Integrate
\(\int \frac{1}{1+x} dx = \ln(1+x)\)
Step 1: Start with geometric series for \(\frac{1}{1+x}\)

From the geometric series: \(\frac{1}{1-u} = \sum_{n=0}^{\infty} u^n\) for \(|u| < 1\)

Substitute \(u = -x\): \(\frac{1}{1-(-x)} = \frac{1}{1+x} = \sum_{n=0}^{\infty} (-x)^n = \sum_{n=0}^{\infty} (-1)^n x^n\)

Step 2: Integrate both sides

\(\int \frac{1}{1+x} dx = \int \sum_{n=0}^{\infty} (-1)^n x^n dx\)

\(\ln(1+x) = \sum_{n=0}^{\infty} \int (-1)^n x^n dx = \sum_{n=0}^{\infty} (-1)^n \frac{x^{n+1}}{n+1} + C\)

Step 3: Determine the constant of integration

At \(x = 0\): \(\ln(1+0) = \ln(1) = 0\)

And the series at \(x = 0\): \(\sum_{n=0}^{\infty} (-1)^n \frac{0^{n+1}}{n+1} = 0\)

Therefore, \(C = 0\)

Step 4: Rewrite the series starting from n=1

\(\ln(1+x) = \sum_{n=0}^{\infty} (-1)^n \frac{x^{n+1}}{n+1} = \sum_{n=1}^{\infty} (-1)^{n-1} \frac{x^n}{n}\)

(Substituted \(m = n+1\), so when \(n = 0\), \(m = 1\))

Step 5: Determine interval of convergence

Since integration doesn't change the radius of convergence, it's still \(R = 1\)

Test endpoints: At \(x = 1\): \(\sum_{n=1}^{\infty} \frac{(-1)^{n-1}}{n}\) (alternating harmonic series - converges)

At \(x = -1\): \(\sum_{n=1}^{\infty} \frac{(-1)^{n-1}(-1)^n}{n} = \sum_{n=1}^{\infty} \frac{-1}{n}\) (harmonic series - diverges)

Power series: \(\ln(1+x) = \sum_{n=1}^{\infty} (-1)^{n-1} \frac{x^n}{n} = x - \frac{x^2}{2} + \frac{x^3}{3} - \frac{x^4}{4} + ...\)
Interval of convergence: \((-1, 1]\)
Final answer:

The power series representation is \(\sum_{n=1}^{\infty} (-1)^{n-1} \frac{x^n}{n}\) with interval of convergence \((-1, 1]\).

Applied rules:

• Term-by-term Integration: Series can be integrated term by term within radius of convergence

• Constant Determination: Use known function value to find integration constant

• Endpoint Testing: Always test endpoints after integration

3 Differentiation of Power Series
Exercise 3
Find a power series representation for:
\(f(x) = \frac{1}{(1-x)^2}\) by differentiating the geometric series.
Definition:

Differentiation of Series: \(\frac{d}{dx} \sum_{n=0}^{\infty} c_n x^n = \sum_{n=1}^{\infty} nc_n x^{n-1}\)

Term-by-term Differentiation: Valid within the radius of convergence

Start with
\(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n\)
Differentiate
\(\frac{d}{dx}\left[\frac{1}{1-x}\right] = \frac{1}{(1-x)^2}\)
Step 1: Start with geometric series

We know: \(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n = 1 + x + x^2 + x^3 + ...\)

Step 2: Differentiate both sides

Left side: \(\frac{d}{dx}\left[\frac{1}{1-x}\right] = \frac{d}{dx}[(1-x)^{-1}] = -(1-x)^{-2} \cdot (-1) = \frac{1}{(1-x)^2}\)

Right side: \(\frac{d}{dx}\left[\sum_{n=0}^{\infty} x^n\right] = \sum_{n=0}^{\infty} \frac{d}{dx}[x^n] = \sum_{n=1}^{\infty} nx^{n-1}\)

(Note: the n=0 term vanishes since \(\frac{d}{dx}[x^0] = \frac{d}{dx}[1] = 0\))

Step 3: Rewrite the differentiated series

\(\sum_{n=1}^{\infty} nx^{n-1}\) can be rewritten by substituting \(m = n-1\), so \(n = m+1\)

When \(n = 1\), \(m = 0\): \(\sum_{m=0}^{\infty} (m+1)x^m = \sum_{n=0}^{\infty} (n+1)x^n\)

Step 4: Write the final result

\(\frac{1}{(1-x)^2} = \sum_{n=0}^{\infty} (n+1)x^n = 1 + 2x + 3x^2 + 4x^3 + ...\)

Step 5: Determine interval of convergence

Differentiation doesn't change the radius of convergence, so \(R = 1\)

At \(x = 1\): The series becomes \(\sum_{n=0}^{\infty} (n+1) = 1 + 2 + 3 + ...\), which diverges

At \(x = -1\): The series becomes \(\sum_{n=0}^{\infty} (n+1)(-1)^n\), which also diverges by the divergence test

Power series: \(\frac{1}{(1-x)^2} = \sum_{n=0}^{\infty} (n+1)x^n = 1 + 2x + 3x^2 + 4x^3 + ...\)
Interval of convergence: \((-1, 1)\)
Final answer:

The power series representation is \(\sum_{n=0}^{\infty} (n+1)x^n\) with interval of convergence \((-1, 1)\).

Applied rules:

• Term-by-term Differentiation: Series can be differentiated term by term within radius of convergence

• Index Shifting: Adjust indices when manipulating series after operations

• Convergence Preservation: Differentiation preserves radius of convergence

Rules and methods, laws,...
\(f(x) = \sum_{n=0}^{\infty} c_n(x-a)^n\)
General Power Series
Geometric
\(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n\)
For |x| < 1
Natural Log
\(\ln(1+x) = \sum_{n=1}^{\infty} (-1)^{n-1}\frac{x^n}{n}\)
For -1 < x ≤ 1
Exponential
\(e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!}\)
For all x
Key definitions:

Power Series: An infinite series of the form \(\sum_{n=0}^{\infty} c_n(x-a)^n\) where \(c_n\) are coefficients and \(a\) is the center

Maclaurin Series: A Taylor series centered at \(a = 0\)

Taylor Series: A power series representation of a function using derivatives: \(f(x) = \sum_{n=0}^{\infty} \frac{f^{(n)}(a)}{n!}(x-a)^n\)

Complete methodology:
  1. Identify the function: Recognize if it's related to known series
  2. Manipulate known series: Use substitution, integration, or differentiation
  3. Apply operations: Carefully perform the required operation
  4. Determine convergence: Find the interval of convergence
  5. Verify results: Check with specific values or derivatives
Tip 1: Always start with the most basic series you know (geometric, exponential).
Tip 2: Integration increases the exponent by 1 and divides by the new exponent.
Tip 3: Differentiation decreases the exponent by 1 and multiplies by the original exponent.
Tip 4: Always check the constant of integration when integrating.
Common errors: Forgetting to adjust the index after differentiation, missing the constant of integration, not checking convergence intervals.
Exam preparation: Memorize key series, practice manipulation techniques, master convergence testing.
Formulas to know by heart:

• Geometric Series: \(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n\) for \(|x| < 1\)

• Natural Log: \(\ln(1+x) = \sum_{n=1}^{\infty} (-1)^{n-1}\frac{x^n}{n}\) for \(-1 < x \leq 1\)

• Exponential: \(e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!}\) for all \(x\)

• Sine: \(\sin x = \sum_{n=0}^{\infty} (-1)^n\frac{x^{2n+1}}{(2n+1)!}\) for all \(x\)

• Cosine: \(\cos x = \sum_{n=0}^{\infty} (-1)^n\frac{x^{2n}}{(2n)!}\) for all \(x\)

Solution: Exercises 4 to 5
4 Substitution Method
Exercise 4
Find a power series representation for:
\(f(x) = \frac{1}{1+x^2}\) and use it to find a series for \(\arctan(x)\).
Definition:

Substitution Method: Replace the variable in a known series with another expression

Arctangent Series: \(\arctan(x) = \int \frac{1}{1+x^2} dx\)

Start with
\(\frac{1}{1-u} = \sum_{n=0}^{\infty} u^n\)
Substitute
\(u = -x^2\)
Step 1: Start with geometric series

We know: \(\frac{1}{1-u} = \sum_{n=0}^{\infty} u^n\) for \(|u| < 1\)

Step 2: Substitute \(u = -x^2\)

\(\frac{1}{1-(-x^2)} = \frac{1}{1+x^2} = \sum_{n=0}^{\infty} (-x^2)^n = \sum_{n=0}^{\infty} (-1)^n x^{2n}\)

Step 3: Determine convergence condition

For convergence: \(|u| < 1 \Rightarrow |-x^2| < 1 \Rightarrow |x^2| < 1 \Rightarrow |x| < 1\)

Step 4: Find series for \(\arctan(x)\) by integration

\(\arctan(x) = \int \frac{1}{1+x^2} dx = \int \sum_{n=0}^{\infty} (-1)^n x^{2n} dx\)

\(= \sum_{n=0}^{\infty} (-1)^n \int x^{2n} dx = \sum_{n=0}^{\infty} (-1)^n \frac{x^{2n+1}}{2n+1} + C\)

Step 5: Determine the constant of integration

At \(x = 0\): \(\arctan(0) = 0\)

And the series at \(x = 0\): \(\sum_{n=0}^{\infty} (-1)^n \frac{0^{2n+1}}{2n+1} = 0\)

Therefore, \(C = 0\)

Step 6: Write final results

\(\frac{1}{1+x^2} = \sum_{n=0}^{\infty} (-1)^n x^{2n} = 1 - x^2 + x^4 - x^6 + ...\)

\(\arctan(x) = \sum_{n=0}^{\infty} (-1)^n \frac{x^{2n+1}}{2n+1} = x - \frac{x^3}{3} + \frac{x^5}{5} - \frac{x^7}{7} + ...\)

For \(\frac{1}{1+x^2}\): \(\sum_{n=0}^{\infty} (-1)^n x^{2n}\) on \((-1, 1)\)
For \(\arctan(x)\): \(\sum_{n=0}^{\infty} (-1)^n \frac{x^{2n+1}}{2n+1}\) on \([-1, 1]\)
Final answer:

\(\frac{1}{1+x^2} = \sum_{n=0}^{\infty} (-1)^n x^{2n}\) and \(\arctan(x) = \sum_{n=0}^{\infty} (-1)^n \frac{x^{2n+1}}{2n+1}\).

Applied rules:

• Substitution: Replace variable in known series with algebraic expression

• Integration: Term-by-term integration preserves convergence properties

• Arctangent Relationship: \(\frac{d}{dx}[\arctan(x)] = \frac{1}{1+x^2}\)

5 Taylor Series
Exercise 5
Find the Taylor series for \(f(x) = e^x\) centered at \(a = 0\) (Maclaurin series) and verify it using the definition.
Definition:

Taylor Series: \(f(x) = \sum_{n=0}^{\infty} \frac{f^{(n)}(a)}{n!}(x-a)^n\)

Maclaurin Series: Taylor series with center \(a = 0\)

Function
\(f(x) = e^x\)
Derivatives
\(f^{(n)}(x) = e^x\) for all n
Step 1: Find all derivatives of \(f(x) = e^x\)

\(f(x) = e^x\)
\(f'(x) = e^x\)
\(f''(x) = e^x\)
\(f'''(x) = e^x\)
\(\vdots\)
\(f^{(n)}(x) = e^x\) for all \(n \geq 0\)

Step 2: Evaluate derivatives at \(a = 0\)

\(f(0) = e^0 = 1\)
\(f'(0) = e^0 = 1\)
\(f''(0) = e^0 = 1\)
\(f'''(0) = e^0 = 1\)
\(\vdots\)
\(f^{(n)}(0) = e^0 = 1\) for all \(n \geq 0\)

Step 3: Apply Taylor series formula

For Maclaurin series (\(a = 0\)): \(f(x) = \sum_{n=0}^{\infty} \frac{f^{(n)}(0)}{n!}x^n\)

\(e^x = \sum_{n=0}^{\infty} \frac{1}{n!}x^n = \sum_{n=0}^{\infty} \frac{x^n}{n!}\)

Step 4: Verify using ratio test

To find radius of convergence: \(\lim_{n \to \infty} \left|\frac{a_{n+1}}{a_n}\right| = \lim_{n \to \infty} \left|\frac{\frac{x^{n+1}}{(n+1)!}}{\frac{x^n}{n!}}\right|\)

\(= \lim_{n \to \infty} \left|\frac{x^{n+1}}{(n+1)!} \cdot \frac{n!}{x^n}\right| = \lim_{n \to \infty} \left|\frac{x}{n+1}\right| = 0\)

Since the limit is 0 < 1 for all finite x, the series converges for all real numbers.

Step 5: Write the explicit series

\(e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!} = \frac{x^0}{0!} + \frac{x^1}{1!} + \frac{x^2}{2!} + \frac{x^3}{3!} + \frac{x^4}{4!} + ...\)

\(= 1 + x + \frac{x^2}{2} + \frac{x^3}{6} + \frac{x^4}{24} + ...\)

Step 6: Verify with known values

At \(x = 1\): \(e^1 = e \approx 2.718\)

Partial sum: \(1 + 1 + \frac{1}{2} + \frac{1}{6} + \frac{1}{24} + \frac{1}{120} = 2.717...\) (close to e)

Taylor series: \(e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!} = 1 + x + \frac{x^2}{2!} + \frac{x^3}{3!} + ...\)
Interval of convergence: \((-\infty, \infty)\)
Final answer:

The Maclaurin series for \(e^x\) is \(\sum_{n=0}^{\infty} \frac{x^n}{n!}\) with infinite radius of convergence.

Applied rules:

• Taylor Series Formula: Uses derivatives at the center point

• Exponential Property: Derivative of \(e^x\) is itself

• Ratio Test: Determines radius of convergence

Comprehensive Summary: Functions as Power Series
\(f(x) = \sum_{n=0}^{\infty} c_n(x-a)^n\)
Taylor Series Formula
Key definitions:

Power Series: An infinite series of the form \(\sum_{n=0}^{\infty} c_n(x-a)^n\) where \(c_n\) are constants called coefficients, and \(a\) is the center of the series.

Radius of Convergence (R): The non-negative number such that the series converges absolutely for \(|x-a| < R\) and diverges for \(|x-a| > R\).

Interval of Convergence: The set of all values of \(x\) for which the power series converges.

Maclaurin Series: A Taylor series centered at \(a = 0\), giving \(\sum_{n=0}^{\infty} \frac{f^{(n)}(0)}{n!}x^n\).

Taylor Series: A power series representation of a function using its derivatives: \(\sum_{n=0}^{\infty} \frac{f^{(n)}(a)}{n!}(x-a)^n\).

Complete methodology for finding power series representations:
  1. Direct Formula (Taylor/Maclaurin): Compute derivatives and apply the formula
  2. Substitution: Replace variables in known series
  3. Algebraic Operations: Add, subtract, multiply, or divide known series
  4. Differentiation: Differentiate known series term by term
  5. Integration: Integrate known series term by term
  6. Long Division: For rational functions
Tip 1: Memorize the basic power series: geometric, exponential, sine, cosine, natural log.
Tip 2: Integration adds 1 to the exponent and divides by the new exponent.
Tip 3: Differentiation subtracts 1 from the exponent and multiplies by the original exponent.
Tip 4: Always determine the interval of convergence after finding the series.
Tip 5: Check your work by evaluating the series at a simple point and comparing with the original function.
Common errors: Forgetting to adjust indices after operations, not checking convergence intervals, making arithmetic mistakes in derivatives.
Key insights: Power series can be manipulated algebraically like polynomials within their radius of convergence.
Essential formulas and properties:

• Basic Geometric: \(\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n\) for \(|x| < 1\)

• Exponential: \(e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!}\) for all \(x\)

• Sine: \(\sin x = \sum_{n=0}^{\infty} (-1)^n\frac{x^{2n+1}}{(2n+1)!}\) for all \(x\)

• Cosine: \(\cos x = \sum_{n=0}^{\infty} (-1)^n\frac{x^{2n}}{(2n)!}\) for all \(x\)

• Natural Log: \(\ln(1+x) = \sum_{n=1}^{\infty} (-1)^{n-1}\frac{x^n}{n}\) for \(-1 < x \leq 1\)

• Differentiation: \(\frac{d}{dx}\left[\sum_{n=0}^{\infty} c_n(x-a)^n\right] = \sum_{n=1}^{\infty} nc_n(x-a)^{n-1}\)

• Integration: \(\int\left[\sum_{n=0}^{\infty} c_n(x-a)^n\right]dx = C + \sum_{n=0}^{\infty} \frac{c_n(x-a)^{n+1}}{n+1}\)

Visual Analysis: Series Approximations
Comparing Function and Series Approximations
Visual representation of how partial sums of power series approximate functions:
\(e^x\) approximation with first N terms
\(\sin x\) approximation with first N terms
\(\frac{1}{1-x}\) approximation with first N terms

Analysis: The chart shows how adding more terms improves the approximation of functions by their power series.

  • Exponential function: Series converges for all x, excellent approximation near origin
  • Sine function: Series converges for all x, periodic nature captured by odd-powered terms
  • Geometric series: Only converges for |x| < 1, divergence outside radius is evident

Questions & Answers

Question: I understand that we can differentiate and integrate power series term by term, but why does this work? What's the theoretical basis for this?

Answer: The ability to differentiate and integrate power series term by term is based on fundamental theorems in analysis:

Theoretical Basis:

  • Uniform Convergence: Within the radius of convergence, power series converge uniformly on compact subsets. This allows us to interchange the order of operations.
  • Term-by-Term Differentiation Theorem: If \(f(x) = \sum_{n=0}^{\infty} c_n(x-a)^n\) converges for \(|x-a| < R\), then \(f'(x) = \sum_{n=1}^{\infty} nc_n(x-a)^{n-1}\) also converges for \(|x-a| < R\).
  • Term-by-Term Integration Theorem: The integral \(\int f(x)dx = C + \sum_{n=0}^{\infty} \frac{c_n(x-a)^{n+1}}{n+1}\) has the same radius of convergence as the original series.

The key insight is that within the radius of convergence, the power series behaves "nicely" enough that calculus operations can be performed term by term. However, at the boundary points of convergence, these operations may not be valid.

For example, while \(\frac{1}{1-x} = \sum x^n\) converges for \(|x| < 1\), differentiating gives \(\frac{1}{(1-x)^2} = \sum nx^{n-1}\), which also converges for \(|x| < 1\), but the behavior at the boundary changes.

Question: When I find a power series representation for a function, how do I know if it's the only one? Could there be multiple different power series representing the same function?

Answer: This is an excellent question about the uniqueness of power series representations!

Uniqueness Theorem: If a function \(f(x)\) can be represented as a power series centered at \(a\), i.e., \(f(x) = \sum_{n=0}^{\infty} c_n(x-a)^n\) for \(|x-a| < R\), then this representation is unique.

More specifically, the coefficients are uniquely determined by the function's derivatives: \(c_n = \frac{f^{(n)}(a)}{n!}\).

Important Implications:

  • If you find any power series representation for a function, it must be the Taylor series of that function.
  • Different methods (substitution, differentiation, integration) will lead to the same series, though they might look different initially.
  • The radius of convergence is also unique for a given function and center point.

For example, whether you derive the series for \(\frac{1}{1+x^2}\) by substitution into the geometric series or by computing derivatives directly, you'll get the same result: \(\sum (-1)^n x^{2n}\).

However, note that the same function might have different power series representations centered at different points, each with their own radius of convergence.

Question: When integrating a power series, I sometimes forget to include the constant of integration. How can I systematically remember to do this, and how do I find its value?

Answer: This is a very common oversight! Here's a systematic approach:

Systematic Approach:

  1. Always write the constant: When you integrate, immediately write "+ C" at the end of the series.
  2. Use a known point: Find a value of x where you know the exact value of the function you're integrating.
  3. Substitute and solve: Plug in the known x-value and solve for C.

Examples of known points:

  • For \(\int \frac{1}{1+x} dx = \ln(1+x)\), use \(x = 0\): \(\ln(1) = 0\), so series at \(x = 0\) must equal 0.
  • For \(\int \frac{1}{1+x^2} dx = \arctan(x)\), use \(x = 0\): \(\arctan(0) = 0\).
  • For definite integrals, you don't need the constant as it cancels out.

Memory aid: Think of integration as "undoing" differentiation. When you differentiate, you lose information about the constant term. When you integrate, you must recover this information using additional information (the known point).

For example, when finding \(\int \sum x^n dx\), immediately write: \(\sum \frac{x^{n+1}}{n+1} + C\), then use the fact that at \(x = 0\), the result should be 0 to find \(C = 0\).

Question: Do differentiation and integration of power series change the radius of convergence? What about the behavior at the endpoints?

Answer: This is a crucial distinction that many students miss!

Radius of Convergence: Differentiation and integration DO NOT change the radius of convergence. If the original series has radius R, then both the differentiated and integrated series have the same radius R.

Endpoint Behavior: However, the behavior at the endpoints CAN change:

  • Integration: May improve convergence at endpoints. A series that diverged at an endpoint might converge after integration.
  • Differentiation: May worsen convergence at endpoints. A series that converged at an endpoint might diverge after differentiation.

Example: Consider \(\ln(1+x) = \sum_{n=1}^{\infty} (-1)^{n-1}\frac{x^n}{n}\) (from integrating \(\sum (-x)^{n-1}\)).

The original geometric series \(\sum (-x)^{n-1}\) diverges at \(x = 1\), but after integration, the series for \(\ln(1+x)\) converges at \(x = 1\) (to \(\ln 2\)).

This is why you must always re-test the endpoints after differentiation or integration to determine the exact interval of convergence for the new series.

Question: How do I decide whether to use direct Taylor series computation or manipulate known series? Is one method preferred over the other?

Answer: The choice depends on the function and context:

Use Direct Taylor Computation when:

  • The function has easily computable derivatives (like \(e^x\), \(\sin x\), \(\cos x\))
  • You need to prove the series exists or understand the theoretical foundation
  • No obvious known series to manipulate exists

Use Series Manipulation when:

  • The function relates to a known series (substitution, integration, differentiation)
  • Derivatives become complex quickly
  • You want a quick result for computational purposes

Strategic Approach:

  1. First, scan for relationships to known series
  2. Consider substitution: Does the function look like a known series with a different input?
  3. Consider operations: Is the function a derivative or integral of something familiar?
  4. Fall back to direct computation if no shortcuts are apparent

For example, for \(\frac{1}{1+x^2}\), substitution into the geometric series is much faster than computing derivatives of this function. But for \(e^{x^2}\), direct computation might be necessary unless you recognize it as substitution.

Both methods are valuable, and the best approach often combines them strategically.