- Write out the terms in a binomial series
- Find the Taylor series for different functions
- Use Taylor series to solve differential equations
- Use Taylor series to evaluate integrals that don’t have elementary antiderivatives
The Binomial Series
The Main Idea
The binomial series generalizes the familiar binomial theorem to handle any real exponent [latex]r[/latex], not just positive integers. This powerful tool lets you find power series for functions like [latex]\sqrt{1+x}[/latex], [latex]\frac{1}{1+x}[/latex], and [latex](1+x)^{-2}[/latex].
The generalized binomial coefficient: For any real number [latex]r[/latex] and non-negative integer [latex]n[/latex]:
[latex](\begin{array}{c}r \ n\end{array}) = \frac{r(r-1)(r-2)\cdots(r-n+1)}{n!}[/latex]
The binomial series formula: [latex](1+x)^r = \sum_{n=0}^{\infty} (\begin{array}{c}r \ n\end{array}) x^n = 1 + rx + \frac{r(r-1)}{2!}x^2 + \frac{r(r-1)(r-2)}{3!}x^3 + \cdots[/latex]
When [latex]r[/latex] is a non-negative integer, the series terminates naturally (since higher-order derivatives become zero), giving you the familiar finite binomial expansion. When [latex]r[/latex] is any other real number, you get an infinite series.
Interval of convergence: The binomial series converges for [latex]|x| < 1[/latex]. Endpoint behavior depends on [latex]r[/latex]:
- [latex]r \geq 0[/latex]: converges at both [latex]x = \pm 1[/latex]
- [latex]-1 < r < 0[/latex]: converges at [latex]x = 1[/latex], diverges at [latex]x = -1[/latex]
- [latex]r < -1[/latex]: diverges at both endpoints
Common applications:
- [latex]\sqrt{1+x} = (1+x)^{1/2}[/latex] with [latex]r = \frac{1}{2}[/latex]
- [latex]\frac{1}{\sqrt{1+x}} = (1+x)^{-1/2}[/latex] with [latex]r = -\frac{1}{2}[/latex]
- [latex]\frac{1}{1+x} = (1+x)^{-1}[/latex] with [latex]r = -1[/latex]
The binomial series is particularly valuable for approximating roots and reciprocals when [latex]x[/latex] is small. You can use just the first few terms to get good approximations, with Taylor’s theorem helping you bound the error.
Find the binomial series for [latex]f\left(x\right)=\frac{1}{{\left(1+x\right)}^{2}}[/latex].
Common Functions Expressed as Taylor Series
The Main Idea
Instead of deriving Taylor series from scratch every time, you can build a powerful toolkit of standard series and use them to find series for more complex functions. This reference approach makes Taylor series much more practical and efficient.
Your essential reference table:
- [latex]\frac{1}{1-x} = \sum_{n=0}^{\infty} x^n[/latex] for [latex]|x| < 1[/latex]
- [latex]e^x = \sum_{n=0}^{\infty} \frac{x^n}{n!}[/latex] for all [latex]x[/latex]
- [latex]\sin x = \sum_{n=0}^{\infty} (-1)^n \frac{x^{2n+1}}{(2n+1)!}[/latex] for all [latex]x[/latex]
- [latex]\cos x = \sum_{n=0}^{\infty} (-1)^n \frac{x^{2n}}{(2n)!}[/latex] for all [latex]x[/latex]
- [latex]\ln(1+x) = \sum_{n=1}^{\infty} (-1)^{n+1} \frac{x^n}{n}[/latex] for [latex]-1 < x \leq 1[/latex]
- [latex]\tan^{-1} x = \sum_{n=0}^{\infty} (-1)^n \frac{x^{2n+1}}{2n+1}[/latex] for [latex]|x| \leq 1[/latex]
- [latex](1+x)^r = \sum_{n=0}^{\infty} \binom{r}{n} x^n[/latex] for [latex]|x| < 1[/latex]
Building new series from old ones: Use substitution, differentiation, integration, and algebraic manipulation:
- For [latex]\cos \sqrt{x}[/latex]: substitute [latex]\sqrt{x}[/latex] into the cosine series
- For [latex]\sinh x = \frac{e^x - e^{-x}}{2}[/latex]: combine exponential series
- For [latex]\frac{1}{\sqrt{1+x}}[/latex]: differentiate the series for [latex]\sqrt{1+x}[/latex]
Key patterns to recognize:
- Exponential-related functions: Use [latex]e^x[/latex] series with appropriate substitutions
- Trigonometric variants: Modify [latex]\sin x[/latex] and [latex]\cos x[/latex] series
- Rational functions: Often relate to geometric series [latex]\frac{1}{1-x}[/latex]
- Root and power functions: Use binomial series latex^r[/latex]
Instead of computing derivatives repeatedly, ask “Which standard series can I modify to get this function?” This approach is faster, less error-prone, and builds your intuition about how different functions relate to each other.
Find the Maclaurin series for [latex]\sin\left({x}^{2}\right)[/latex].
Find the binomial series for [latex]f\left(x\right)=\frac{1}{{\left(1+x\right)}^{\frac{3}{2}}}[/latex]
Solving Differential Equations with Power Series
The Main Idea
Power series offer a systematic approach to solving differential equations that can’t be handled by separation of variables, integrating factors, or other elementary techniques. The method transforms a differential equation into a problem about finding coefficients in a power series.
Assume your solution has the form [latex]y = \sum_{n=0}^{\infty} c_n x^n[/latex], then use the differential equation to determine what the coefficients [latex]c_n[/latex] must be.
Problem-Solving Strategy:
- Assume a power series solution: [latex]y = c_0 + c_1 x + c_2 x^2 + c_3 x^3 + \cdots[/latex]
- Differentiate term-by-term: [latex]y' = c_1 + 2c_2 x + 3c_3 x^2 + \cdots[/latex]
- Substitute into the differential equation: Replace [latex]y[/latex], [latex]y'[/latex], etc. with their series
- Match coefficients: Use the uniqueness of power series to equate coefficients of like powers
- Apply initial conditions: Use [latex]y(0)[/latex], [latex]y'(0)[/latex], etc. to find specific coefficient values
The uniqueness of power series representations means that if two power series are equal, their corresponding coefficients must be equal. This gives you a system of equations to solve for the coefficients.
Often you’ll find that [latex]c_{n+k} = f(c_n)[/latex] for some function [latex]f[/latex]. This creates patterns where coefficients can be expressed in terms of just a few initial coefficients.
Power series solutions are especially valuable for:
- Linear differential equations with polynomial coefficients
- Equations where elementary methods fail (like Airy’s equation [latex]y'' - xy = 0[/latex])
- Problems requiring solutions near specific points
Sometimes your coefficient pattern will match a known Taylor series (like [latex]e^x[/latex]), giving you a closed-form solution. Other times, the power series itself is the most useful form of the solution.
Use power series to solve [latex]{y}^{\prime }=2y,y\left(0\right)=5[/latex].
Use power series to solve [latex]y^{\prime\prime}+{x}^{2}y=0[/latex] with the initial condition [latex]y\left(0\right)=a[/latex] and [latex]{y}^{\prime }\left(0\right)=b[/latex].
Evaluating Nonelementary Integrals
The Main Idea
Some integrals simply can’t be solved using the integration techniques you’ve learned. Functions like [latex]e^{-x^2}[/latex], [latex]\frac{\sin x}{x}[/latex], and [latex]\frac{1}{\sqrt{1-k^2\sin^2\theta}}[/latex] have antiderivatives that aren’t expressible as elementary functions. Power series provide the key to unlocking these “impossible” integrals.
When you can’t integrate directly, represent the integrand as a power series, then integrate term by term. This transforms an unsolvable integral into a manageable infinite series.
Problem-Solving Strategy:
- Express the integrand as a power series using known series or substitutions
- Integrate term by term – power series can be integrated just like polynomials
- Evaluate definite integrals by substituting limits into your series
- Estimate accuracy using alternating series test or other convergence tools
Power series don’t just solve differential equations – they extend your integration capabilities to a vast class of functions that would otherwise be inaccessible. When traditional integration fails, power series succeed by transforming the problem into polynomial arithmetic.
Express [latex]\displaystyle\int \cos\sqrt{x}dx[/latex] as an infinite series. Evaluate [latex]{\displaystyle\int }_{0}^{1}\cos\sqrt{x}dx[/latex] to within an error of [latex]0.01[/latex].
Use the first five terms of the Maclaurin series for [latex]{e}^{\text{-}\frac{{x}^{2}}{2}}[/latex] to estimate the probability that a randomly selected test score is between [latex]100[/latex] and [latex]150[/latex]. Use the alternating series test to determine the accuracy of this estimate.