AP Calculus BC

Common Maclaurin Series Table (BC)

The four Maclaurin series to memorize are e^x = sum of x^n/n! (all x), sin x = sum of (-1)^n x^(2n+1)/(2n+1)! (all x), cos x = sum of (-1)^n x^(2n)/(2n)! (all x), and 1/(1-x) = sum of x^n (only for |x| < 1). Two common extensions, ln(1+x) and arctan x, hold for -1 < x <= 1 and |x| <= 1.

The four series the CED names

All of Unit 10 is BC-only, and these four Maclaurin series are the ones the College Board builds on directly in Topic 10.14: exe^x, sinx\sin x, and cosx\cos x are named as the foundation for constructing other series (LIM-8.F.2), and 11x\frac{1}{1-x} is the geometric building block (LIM-8.F.1). Memorize these four cold. Almost every other series on the BC exam is reached by reshaping one of them.

FunctionMaclaurin series (general term)Converges forWhen to reach for it
exe^xn=0xnn!=1+x+x22!+x33!+\displaystyle\sum_{n=0}^{\infty} \frac{x^n}{n!} = 1 + x + \frac{x^2}{2!} + \frac{x^3}{3!} + \cdotsall real xxAny e(expression)e^{(\text{expression})}: substitute the exponent for xx. Converges everywhere, so there is no interval to check.
sinx\sin xn=0(1)nx2n+1(2n+1)!=xx33!+x55!\displaystyle\sum_{n=0}^{\infty} \frac{(-1)^n x^{2n+1}}{(2n+1)!} = x - \frac{x^3}{3!} + \frac{x^5}{5!} - \cdotsall real xxOdd powers only, matching that sin\sin is an odd function. Use for sin\sin of any expression; safe to substitute freely.
cosx\cos xn=0(1)nx2n(2n)!=1x22!+x44!\displaystyle\sum_{n=0}^{\infty} \frac{(-1)^n x^{2n}}{(2n)!} = 1 - \frac{x^2}{2!} + \frac{x^4}{4!} - \cdotsall real xxEven powers only, matching that cos\cos is even. It is the term-by-term derivative of the sinx\sin x series, so you can recover one from the other.
11x\frac{1}{1-x}n=0xn=1+x+x2+x3+\displaystyle\sum_{n=0}^{\infty} x^n = 1 + x + x^2 + x^3 + \cdotsx<1|x| < 1The geometric series with ratio xx (LIM-8.F.1). Base for 11+x\frac{1}{1+x}, 11x2\frac{1}{1-x^2}, and similar rationals. Diverges once x1|x| \ge 1, so always carry the interval.

The three that converge for all xx (exe^x, sinx\sin x, cosx\cos x) are the ones you can substitute into without a second thought: any expression can replace xx and the result still converges everywhere. The geometric series is the exception. It collapses the moment x1|x| \ge 1, so any substitution into it comes with an interval condition attached.

Two extensions worth memorizing

These two are not in the CED (Course and Exam Description) essential knowledge, so treat them as commonly-taught extras rather than required facts. You can derive each by integrating a geometric-type series term by term (Topic 10.15), but memorizing them saves time. They are labeled here as extensions so you know they sit outside the core four above.

Function (extension)Maclaurin series (general term)Converges forWhere it comes from
ln(1+x)\ln(1+x)n=1(1)n1xnn=xx22+x33\displaystyle\sum_{n=1}^{\infty} \frac{(-1)^{n-1} x^n}{n} = x - \frac{x^2}{2} + \frac{x^3}{3} - \cdots1<x1-1 < x \le 1Integrate the 11+x\frac{1}{1+x} series term by term. The sum starts at n=1n = 1: there is no constant term because ln1=0\ln 1 = 0.
arctanx\arctan xn=0(1)nx2n+12n+1=xx33+x55\displaystyle\sum_{n=0}^{\infty} \frac{(-1)^n x^{2n+1}}{2n+1} = x - \frac{x^3}{3} + \frac{x^5}{5} - \cdots1x1-1 \le x \le 1Integrate the 11+x2\frac{1}{1+x^2} series term by term. Odd powers like sinx\sin x, but the denominators are 2n+12n+1, not factorials.

Both extensions come from integrating a known series, which is why their intervals pick up an endpoint the plain geometric series does not have. Term-by-term integration keeps the radius of convergence the same (LIM-8.D.6) but can change what happens at the endpoints, so ln(1+x)\ln(1+x) converges at x=1x = 1 and arctanx\arctan x converges at both x=±1x = \pm 1. An interval question still needs each endpoint tested on its own (Topic 10.13).

Choosing which series to start from

Method selection is the real skill. Before computing anything, look at f(x)f(x) and decide which base series it is a disguised version of. The derivative formula f(n)(a)n!\frac{f^{(n)}(a)}{n!} (Topic 10.11) always works but is slow, so save it for functions that match nothing on the list. This table maps a function's shape to the series and the single operation that builds it fastest.

If f(x)f(x) looks likeStart fromDo thisExample
A standard function with xx replaced by an expression: cos(3x)\cos(3x), ex2e^{-x^2}the matching base seriessubstitute the expression in place of xxex2=n=0(1)nx2nn!e^{-x^2} = \displaystyle\sum_{n=0}^{\infty} \frac{(-1)^n x^{2n}}{n!}
A base series times a power of xx: x2sinxx^2 \sin xthe base seriesmultiply every term by that powerx2sinx=n=0(1)nx2n+3(2n+1)!x^2 \sin x = \displaystyle\sum_{n=0}^{\infty} \frac{(-1)^n x^{2n+3}}{(2n+1)!}
A rational form 11±ax\frac{1}{1 \pm ax} or 11±ax2\frac{1}{1 \pm ax^2}11x\frac{1}{1-x}substitute, factoring out constants first11+x2=n=0(1)nx2n\frac{1}{1+x^2} = \displaystyle\sum_{n=0}^{\infty} (-1)^n x^{2n}
A function whose derivative is a known series: ln(1+x)\ln(1+x), arctanx\arctan xthe derivative's series (e.g. 11+x\frac{1}{1+x} or 11+x2\frac{1}{1+x^2})integrate term by term, then fix the constant11+x2dx=arctanx\int \frac{1}{1+x^2}\,dx = \arctan x
Nothing on the list, or a center a0a \ne 0the derivative formulacompute f(n)(a)n!\frac{f^{(n)}(a)}{n!} and attach to (xa)n(x-a)^nlnx\ln x centered at a=1a = 1

You can chain these operations: for x2cos(2x)x^2 \cos(2x), substitute 2x2x into the cosine series, then multiply every term by x2x^2, giving x2cos(2x)=n=0(1)n22nx2n+2(2n)!x^2 \cos(2x) = \displaystyle\sum_{n=0}^{\infty} \frac{(-1)^n 2^{2n} x^{2n+2}}{(2n)!}. The one case with no shortcut is a series centered at a0a \ne 0, such as lnx\ln x at x=1x = 1; there you compute derivatives at aa and attach each coefficient to (xa)n(x-a)^n, not xnx^n. The Taylor and Maclaurin series guide works that example in full.

Watch-outs when using the table

A handful of slips account for most lost points once you start pulling series from this table.

  • Factorials belong to exe^x, sinx\sin x, and cosx\cos x only. The geometric series and its integrals (11x\frac{1}{1-x}, ln(1+x)\ln(1+x), arctanx\arctan x) have plain integer denominators or none at all.
  • Signs travel through a substitution: (x2)n=(1)nx2n(-x^2)^n = (-1)^n x^{2n}, so replacing xx with x2-x^2 changes both the sign pattern and the powers.
  • Intervals do not carry over unchanged. Substituting u=g(x)u = g(x) into a series valid for u<1|u| < 1 gives the condition g(x)<1|g(x)| < 1, which you still have to solve for xx.
  • Mind the starting index. The ln(1+x)\ln(1+x) series starts at n=1n = 1 (its first term is xx), while exe^x, cosx\cos x, and 11x\frac{1}{1-x} start at n=0n = 0.