MathematicsCore22 min read

Sequences and Series

Patterns with a rule, and what happens when you add them all up

This topic appears in:

01

A sequence is a list; a series is its total

A sequence is an ordered list of numbers, each called a term and written a₁, a₂, a₃, … The rule producing them may be given explicitly — "aₙ = 3n − 1" — or recursively, as a starting value plus an instruction for getting from one term to the next.

A series is what you get when you add the terms of a sequence. Sₙ means the sum of the first n terms, and the distinction matters: a question asking for the tenth term and one asking for the sum of the first ten are different questions with different formulae.

TypeHow each term is builtnth term
Arithmetic (AP)add a fixed daₙ = a + (n − 1)d
Geometric (GP)multiply by a fixed raₙ = a rⁿ⁻¹
Harmonic (HP)the reciprocals form an APtake reciprocals, use the AP rule, invert back

Choose Geometric and pull r below 1. The bars shrink towards nothing and the running-total line flattens out — that flattening is the sum to infinity, and it only happens when |r| < 1.

02

Arithmetic progressions

Everything about an AP follows from one number: the common difference d = aₙ − aₙ₋₁, which must be the same between every consecutive pair. Testing that is how you prove a sequence is arithmetic.

The sum has two equivalent forms, and choosing the right one saves a line. Use the first when you know a and d; use the second when you know the first and last terms.

aₙ = a + (n − 1)dSₙ = (n/2)[2a + (n − 1)d]Sₙ = (n/2)(a + l)(l = the last term)arithmetic mean of a and b = (a + b)/2the second sum formula is just the first with l = a + (n−1)d substituted in
a
the first term
d
the common differencecan be negative
n
the number of termsa positive integer

Where Sₙ = (n/2)(a + l) comes from

Write the sum forwards and then backwards underneath it and add the two rows. Every column totals a + l, and there are n columns, so twice the sum is n(a + l). The story that Gauss found this at school may or may not be true; the trick certainly works, and it is worth knowing rather than memorising.

03

Geometric progressions, and when a series can be infinite

A GP has a constant common ratio r = aₙ / aₙ₋₁. Test a sequence by dividing consecutive terms, not by subtracting them.

The most interesting result in the chapter is that an infinite series can have a finite total. If |r| < 1 each term is a fraction of the one before, the terms shrink towards zero fast enough that the running total settles on a limit, and the sum to infinity exists. If |r| ≥ 1 the terms do not shrink and the total grows without bound — there is no sum to infinity, and saying so is the answer.

aₙ = a rⁿ⁻¹Sₙ = a(1 − rⁿ) / (1 − r)(r ≠ 1)S∞ = a / (1 − r)only when |r| < 1geometric mean of a and b = √(ab)the condition |r| < 1 is part of the formula, not a footnote to it
Worked example

The third term of a GP is 12 and the sixth is 96. Find a, r, and the sum of the first eight terms.

  1. ar² = 12 and ar⁵ = 96.Write both given terms using aₙ = arⁿ⁻¹ before doing anything else.
  2. Divide the second by the first: r³ = 96/12 = 8, so r = 2.Dividing eliminates a in one step. Subtracting would not, which is why division is the move for a GP.
  3. Substitute back: a(4) = 12, so a = 3.
  4. S₈ = 3(1 − 2⁸)/(1 − 2) = 3(1 − 256)/(−1) = 765.With r > 1 it is tidier to use a(rⁿ − 1)/(r − 1), which avoids two negatives.

a = 3, r = 2, S₈ = 765

04

Means, and the standard trap

Inserting means between two numbers is a standard question. To insert k arithmetic means between a and b, treat the whole thing as an AP with k + 2 terms, first term a and last term b, and find d. For k geometric means, do the same with a GP and find r.

For any two positive numbers, the arithmetic mean is always greater than or equal to the geometric mean, with equality only when the two numbers are the same. That result — AM ≥ GM — is examinable in its own right.

Count the terms, not the means

Inserting three arithmetic means between 2 and 18 gives a sequence of five terms, not three. So 18 = 2 + 4d and d = 4, giving 2, 6, 10, 14, 18. Using n = 3 instead of n = 5 is the single commonest error in this chapter and loses the whole question.

05

Sigma notation, and the standard sums

Σ is shorthand for "add these up". Σ(from r=1 to n) aᵣ means substitute r = 1, 2, 3 … n into the expression and total the results. Three standard sums are quoted rather than derived, and questions combine them.

Σ r = n(n + 1)/2Σ r² = n(n + 1)(2n + 1)/6Σ r³ = [n(n + 1)/2]²Σ (ar + b) = a Σr + bnnote that Σr³ is the square of Σr — a result worth remembering for the check it gives you

Before you leave this chapter

  1. AP: constant difference, aₙ = a + (n−1)d, Sₙ = (n/2)[2a + (n−1)d].
  2. GP: constant ratio, aₙ = arⁿ⁻¹, Sₙ = a(1 − rⁿ)/(1 − r).
  3. S∞ = a/(1 − r) exists only when |r| < 1 — state the condition every time.
  4. Divide consecutive terms to test for a GP; subtract them to test for an AP.
  5. Inserting k means makes a sequence of k + 2 terms.
06

Deciding which progression a word problem is

Applied questions rarely name the progression. The wording does it for you, and the distinction is always the same: does the quantity change by a fixed amount each step, or by a fixed factor?

The question saysProgressionBecause
salary rises by Rs 2000 each yearAPa fixed amount is added
salary rises by 5% each yearGPa fixed multiplier of 1.05
a stack with one fewer brick per layerAPa constant difference
a population doubling every hourGPr = 2
a ball rebounding to ¾ of its heightGPr = 0.75
simple interestAPthe same interest added each period
compound interestGPinterest on the growing total

A percentage change is always geometric

An increase of 8% multiplies by 1.08; a decrease of 8% multiplies by 0.92. Both are constant ratios, so the sequence is geometric even though the wording says "increases by". Treating a percentage change as a common difference is the most frequent misreading in the applied questions.

07

The binomial expansion for a positive integer power

Expanding (a + b)ⁿ by multiplying out is fine for n = 2 and unbearable by n = 7. The binomial theorem writes the answer directly, because the coefficients follow a pattern that can be computed rather than discovered.

Each term is built the same way: the powers of a count down from n, the powers of b count up from 0, and every term has powers summing to n. The coefficient of each term counts how many ways that combination of a's and b's can be chosen from the n brackets — which is exactly what ⁿCr means.

(a + b)ⁿ = ⁿC₀aⁿ + ⁿC₁aⁿ⁻¹b + ⁿC₂aⁿ⁻²b² + … + ⁿCₙbⁿgeneral term: ⁿCr · aⁿ⁻ʳ · bʳⁿCr = n! / (r!(n − r)!)powers of a fall, powers of b rise, and they always total n
n
the powerany positive integer
r
which termcounts from 0, so term r+1
ⁿCr
the coefficientn! / (r!(n − r)!)
Worked example

Find the coefficient of x³ in the expansion of (2 + 3x)⁵.

  1. Identify a = 2, b = 3x, n = 5, and the term needed is r = 3.x³ comes from b³, and b carries the x, so r is 3.
  2. General term = ⁵C₃ · 2⁵⁻³ · (3x)³.Substituting into the general term rather than writing out all six terms.
  3. = 10 × 4 × 27x³.⁵C₃ = 10, 2² = 4, and (3x)³ = 27x³ — the 3 is cubed as well as the x, which is the step most often missed.
  4. = 1080x³, so the coefficient is 1080.The question asks for the coefficient, so the answer is the number alone, without the x³.

1080

Cube the whole term, not just the x

In (3x)³ the bracket means 27x³, not 3x³. Whenever the term in the bracket has a coefficient, raising it to a power raises that coefficient too. Writing the term in brackets before expanding it makes this hard to forget — and it is the single most common error in binomial questions.

Practice questions

6 questions · 20 marks · full working on every one

Try each one on paper first, then open the working. The marks are shown where they are actually awarded, because that is where they are actually lost.

Short questions

3 · 6 marks

Two marks each, in the style of the short-question section of the paper. Answer in two or three lines.

SQ1[2 marks]
Distinguish between a sequence and a series.
Model answer

A sequence is an ordered list of terms, such as 2, 5, 8, 11. A series is the sum of the terms of a sequence, such as 2 + 5 + 8 + 11 = 26.

Examiner tip. Give a short numerical example of each. It makes the distinction unmistakable and takes one line.

SQ2[2 marks]
Find the 20th term of the AP 7, 11, 15, …
Model answer

a = 7 and d = 4, so a₂₀ = 7 + 19(4) = 7 + 76 = 83.

Examiner tip. The multiplier is n − 1, not n. Using 20 × 4 gives 87 and is the classic off-by-one in this chapter.

SQ3[2 marks]
State the condition for an infinite geometric series to have a sum, and give the formula.
Model answer

The sum to infinity exists only when |r| < 1, and then S∞ = a / (1 − r).

Examiner tip. The condition is worth a mark by itself. Quoting the formula without it routinely scores half.

Solved numericals

2 · 8 marks

Full working, one step per line, with the marks shown where they are awarded.

N1[4 marks]
The sum of the first 10 terms of an AP is 155 and the sum of the first 20 is 610. Find a and d.
Full working
  1. S₁₀ = 5[2a + 9d] = 155, so 2a + 9d = 31divide through by 5 immediately to keep the numbers small[1]
  2. S₂₀ = 10[2a + 19d] = 610, so 2a + 19d = 61[1]
  3. Subtracting: 10d = 30, so d = 3the 2a terms cancel[1]
  4. 2a + 27 = 31, so a = 2check: S₁₀ = 5(4 + 27) = 155 ✓[1]

a = 2, d = 3

Examiner tip. Two sums always give two simultaneous equations in a and d. Simplify each one before subtracting and the elimination is a single line.

N2[4 marks]
Express the recurring decimal 0.4747… as a fraction by summing an infinite geometric series.
Full working
  1. Writes it as 0.47 + 0.0047 + 0.000047 + …each block of two digits is one term[1]
  2. Identifies a = 0.47 and r = 0.01each term is the previous one divided by 100[1]
  3. |r| < 1, so S∞ = 0.47 / (1 − 0.01) = 0.47 / 0.99the condition should be stated[1]
  4. = 47/99[1]

47/99

Examiner tip. This is the same result the algebraic 100x − x method gives, and it is a good check. Every recurring decimal is an infinite GP with r a power of 1/10 — which is why every one of them is rational.

Long questions

1 · 6 marks

Theory and numerical together, as they appear in the long-question section.

LQ1[6 marks]
A ball is dropped from a height of 8 m. After each bounce it rises to three quarters of its previous height.
  1. Find the height it reaches after the fourth bounce.
  2. Find the total distance travelled before it comes to rest.
  3. Explain why the total distance is finite even though the ball bounces infinitely often.
Mark scheme
  1. Heights form a GP with a = 8 and r = 0.75; after the 4th bounce the height is 8(0.75)⁴the drop height is the term before the first bounce[1]
  2. = 2.53 m (3 s.f.)[1]
  3. Total distance = the initial 8 m drop + twice each subsequent rise, since each bounce goes up and comes back downthe factor of 2 is the mark most often missed[1]
  4. Rises form a GP with first term 8(0.75) = 6 and r = 0.75, so their sum is 6/(1 − 0.75) = 24[1]
  5. Total = 8 + 2(24) = 56 m[1]
  6. Because |r| < 1, the heights shrink geometrically towards zero fast enough for the infinite sum to converge to a finite limitthe explanation must refer to |r| < 1[1]

(a) 2.53 m (b) 56 m (c) the heights form a convergent GP since |r| < 1

Examiner tip. The bouncing-ball question appears in some form every year. The two things to get right are doubling the rises and not doubling the initial drop.