Lesson 7 of 8 · 18 min

At least, at most and the binomial mean

Translate at most, at least, no more than, no less than and between into inequalities, calculate cumulative binomial probabilities with and without technology, and use the mean npnp and variance np(1−p)np(1 - p).

What you will be able to do

  • Translate the language of probability (at most, at least, no more than, no less than, inclusive, between) into inequalities for XX.
  • Calculate cumulative binomial probabilities by hand with the complement, and with a calculator.
  • Calculate E⁡(X)=np\E(X) = np, Var⁡(X)=np(1−p)\Var(X) = np(1 - p) and the standard deviation, and work back to nn and pp.

Exam questions rarely ask for P(X=r)P(X = r) alone. They ask for the probability that "at least three" seeds germinate or "no more than two" parcels arrive late. Each phrase is an inequality, and for a whole-number variable the boundary is exactly where marks are lost.

Words and symbols for a random variable XX that takes whole-number values.
In wordsIn symbolsValues included (for Bin(12,p)\mathrm{Bin}(12, p))
at most 3, no more than 3X≤3X \le 30, 1, 2, 3
at least 3, no less than 3, no fewer than 3X≥3X \ge 33, 4, …, 12
fewer than 3, less than 3X<3X < 3, which is X≤2X \le 20, 1, 2
more than 3X>3X > 3, which is X≥4X \ge 44, 5, …, 12
between 2 and 5 inclusive2≤X≤52 \le X \le 52, 3, 4, 5
between 2 and 5 (not inclusive)2<X<52 < X < 53, 4

Note

Read "between" carefully

"Between 2 and 5" on its own is ambiguous. QCAA questions normally say "inclusive" when the ends are included. If a question doesn't say, state how you have read it.

Which inequality means "no fewer than 4 successes"?

Choose one answer

By hand, add the probabilities of the values included. When there are many values, use the complement so that there are fewer terms. For X∼Bin(12,0.3)X \sim \mathrm{Bin}(12, 0.3),

P(X≥3)=1−P(X≤2)=1−[P(X=0)+P(X=1)+P(X=2)].P(X \ge 3) = 1 - P(X \le 2) = 1 - \big[P(X = 0) + P(X = 1) + P(X = 2)\big].

With technology, use the binomial cumulative function, which gives P(X≤r)P(X \le r) or, on some calculators, P(a≤X≤b)P(a \le X \le b) for bounds you enter.

Interactive

Binomial probabilities

The shaded bars are P(2≤X≤5)P(2 \le X \le 5). Drag the bounds, or change nn and pp and watch the shape and the mean move.

Worked example

Cumulative binomial probabilities with technology

  • Complex familiar
  • Technology active

At a bakery, 30% of customers buy a meat pie, independently of each other. Of the next 12 customers, let XX be the number who buy a meat pie. Give answers correct to 4 decimal places.

(a) Determine the probability that at least 3 buy a meat pie.

(b) Determine the probability that fewer than 5 buy a meat pie.

5 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Define XX and state X∼Bin(n,p)X \sim \mathrm{Bin}(n, p).

  2. Translate the words into an inequality, then into the form P(X≤r)P(X \le r).

  3. Use the complement for "at least", and evaluate with binomial cdf.

Watch out for

  • Calculating 1−P(X≤3)1 - P(X \le 3) for "at least 3".

  • Including X=5X = 5 in "fewer than 5".

Without technology

Write the inequality, decide whether the values included or the complement need fewer terms, and use P(X=r)=(nr)pr(1−p)n−rP(X = r) = \binom{n}{r} p^r (1 - p)^{n - r} for each. Paper 1 keeps nn small and pp simple, so the arithmetic stays exact.

With technology

Binomial pdf gives P(X=r)P(X = r). Binomial cdf gives P(X≤r)P(X \le r), or P(a≤X≤b)P(a \le X \le b) on calculators that take a lower and an upper bound. Convert the question to these forms first: P(X>4)=1−P(X≤4)P(X > 4) = 1 - P(X \le 4), and P(2<X<6)=P(3≤X≤5)P(2 < X < 6) = P(3 \le X \le 5).

X∼Bin(12,0.3)X \sim \mathrm{Bin}(12, 0.3). Use technology to find the probability that XX is between 2 and 5 inclusive, correct to 4 decimal places.

Round to 4 decimal places.

A binomial variable is a sum of nn independent Bernoulli variables, one for each trial, each with mean pp and variance p(1−p)p(1 - p). Means add, and for independent trials variances add too, which gives the binomial mean and variance.

Key idea

Mean and variance of a binomial distribution

If X∼Bin(n,p)X \sim \mathrm{Bin}(n, p), then

E⁡(X)=np,Var⁡(X)=np(1−p),σ=np(1−p).\E(X) = np, \qquad \Var(X) = np(1 - p), \qquad \sigma = \sqrt{np(1 - p)}.

About 20% of people who receive a text survey reply to it. The survey is sent to 50 people. Let XX be the number who reply. Find the standard deviation of XX, correct to 2 decimal places.

Round to 2 decimal places.

Worked example

Find n and p from the mean and variance

  • Complex familiar
  • Technology free

A binomial random variable XX has mean 6 and variance 4.2. Determine the values of nn and pp, and hence P(X=0)P(X = 0) in index form.

4 steps

Try it yourself first, then check the working one step at a time.

Method

  1. 1−p=Var⁡(X)E⁡(X)1 - p = \dfrac{\Var(X)}{\E(X)}.

  2. n=E⁡(X)pn = \dfrac{\E(X)}{p}, which must be a whole number.

Watch out for

  • Dividing the other way, giving 1−p=64.2>11 - p = \frac{6}{4.2} > 1.

  • Stopping at 1−p=0.71 - p = 0.7 and reporting p=0.7p = 0.7.

Common mistake

The boundary in a complement

P(X≥3)=1−P(X≤2)P(X \ge 3) = 1 - P(X \le 2), not 1−P(X≤3)1 - P(X \le 3). The complement of "3 or more" is "2 or fewer". Write both events out in full before you subtract.

Exam tip

Show the calculator input

On Paper 2 write the model and the probability statement, for example "X∼Bin(12,0.3)X \sim \mathrm{Bin}(12, 0.3), P(X≥3)=1−P(X≤2)=0.7472P(X \ge 3) = 1 - P(X \le 2) = 0.7472". Don't write calculator syntax such as binomcdf as your working.

Summary

  • At most is ≤\le, at least and no less than are ≥\ge; fewer than is <<, more than is >>; between … inclusive includes the ends.

  • For a whole-number variable, P(X<a)=P(X≤a−1)P(X < a) = P(X \le a - 1) and P(X≥a)=1−P(X≤a−1)P(X \ge a) = 1 - P(X \le a - 1).

  • By hand, use the complement to reduce the number of terms. With technology, use binomial pdf and cdf.

  • For X∼Bin(n,p)X \sim \mathrm{Bin}(n, p): E⁡(X)=np\E(X) = np, Var⁡(X)=np(1−p)\Var(X) = np(1 - p) and σ=np(1−p)\sigma = \sqrt{np(1 - p)}.

  • Given the mean and variance, 1−p=Var⁡(X)E⁡(X)1 - p = \dfrac{\Var(X)}{\E(X)}, then n=E⁡(X)pn = \dfrac{\E(X)}{p}.

Finished the lesson?

Try the 3 quick checks above and the lesson completes itself, or mark it complete now.

Lesson 7 of 80% read

Save your progress with a free account