In Unit 3 your random variables took separate values, and you added probabilities from a table. Now the variable can take any value in an interval, like a time or a mass, so you integrate instead of adding. If you are comfortable with definite integrals from Topic 1, most of this topic is applying them carefully. Tick off each skill once you can do it without help.
Skills in this topic
General continuous random variables (about 6 hours). Only tick a skill when you could do it in an exam.
Part 1
Probability is area
Start with data. When you measure a continuous variable, such as how long people wait in a queue, you group the results into class intervals and draw a histogram. Scale it so each bar's area is the relative frequency of that class. The total area is then 1, and the area over an interval estimates the probability of landing in it. Collect more data and narrow the classes, and the tops of the bars settle into a smooth curve. That curve is a probability density function, or PDF.
For a PDF $f$, the probability that $X$ lands between $a$ and $b$ is the area under the curve between them:
$$P(a\le X\le b)=\int_{a}^{b}f(x)\,dx$$
A function has to pass two checks to be a PDF:
1It is never negative, $f(x)\ge0$, because probability cannot be negative.
2The total area under it is 1, because all the probabilities together add to 1.
QCAA write PDFs in two pieces, the rule on its interval and zero everywhere else:
For this PDF, $P(1\le X\le2)=\int_{1}^{2}\frac{1}{8}x\,dx=\Big[\frac{x^{2}}{16}\Big]_{1}^{2}=\frac{4}{16}-\frac{1}{16}=\frac{3}{16}$.
Two things that follow from probability being area. The probability of any single exact value is zero, $P(X=a)=0$, so $P(X<a)$ and $P(X\le a)$ are the same. And the area from the bottom of the interval up to $x$ has its own name, the cumulative distribution function, $F(x)=P(X\le x)$. You will use that idea in Part 4.
Try it yourself
Move the bounds, watch the area
This is the PDF $f(x)=\frac{1}{8}x$ on $0\le x\le4$. Move $a$ and $b$, and the shaded area is $P(a\le X\le b)$. An interval of the same width holds more probability on the right, where the curve is higher. Stretch it across the whole domain and you get 1.
Practise
Exam questions: finding a probability
2023 Paper 1, Question 10
Multiple choice · Technology-free · [1 mark]
The continuous random variable $Y$ has the probability density function
The odd-looking upper limit, $\sqrt{3}-1$, is only there to make the total area 1. It does not affect this integral. (B) comes from subtracting $\frac{1}{8}$ instead of adding it.
Answer
(C) $\frac{5}{8}$
2020 Paper 1, Question 7
Multiple choice · Technology-free · [1 mark]
The life expectancy (in years) of an electronic component can be represented by the probability density function
On Paper 2 I would usually let the calculator's solver do this. On Paper 1, or when the question says "show that", you do it by hand like this.
Game plan: An unknown constant
1Write the integral over the whole interval where the PDF is not zero, and set it equal to 1.
2Integrate, keeping the constant as a letter, then solve for it. On Paper 2, a solver is fine.
3Put the value back into the PDF before you answer any later parts.
Watch the area become 1
Practise
Exam question: an unknown constant
For the written-answer questions, the solutions show where each mark goes, based on the QCAA marking guides. Many marks allow follow-through: if an earlier step is wrong but you use your wrong answer correctly, you can still get the later marks. That is another reason to show every step.
2022 Paper 2, Question 16
Complex familiar · Technology-active · [4 marks]
The time spent waiting in a queue at a certain supermarket is given by $(X+11)$ minutes, where $X$ is a random variable with the probability density function
1 markThe integral equation, total area equal to 1.
1 markThe value of $a$.
Next, translate the question into $X$. The wait is $X+11$, so a wait between 10 and 12 minutes means $10\le X+11\le12$, which is $-1\le X\le1$. This is the step most people miss.
These work exactly like the discrete versions from Unit 3, with integrals in place of sums. For the mean, multiply the PDF by $x$ and integrate over the interval:
$$E(X)=\mu=\int x\,f(x)\,dx$$
The formula sheet gives the variance as $\int(x-\mu)^{2}f(x)\,dx$. I find this version quicker. It is not on the formula sheet, so learn it:
On your calculator, type the cube root as $12^{1/3}$ in the upper bound. The answer has to sit inside the interval, which runs from 0 to about 2.29, and 1.72 does.
Answer
(A) 1.72
2022 Paper 2, Question 5
Multiple choice · Technology-active · [1 mark]
The continuous random variable $X$ has the probability density function
(A) is the variance. The question asks for the standard deviation.
Answer
(B) 0.684
2021 Paper 2, Question 13
Simple familiar · Technology-active · [7 marks]
The amount of gravel (in tonnes) sold by a construction company in a given week is a continuous random variable $X$ and has a probability density function defined by:
Determine the mean time people spend viewing the photo.
The question does not mention $k$, but you cannot find the mean without it. That is what makes this one unfamiliar. Find $k$ first, from the total area:
The $n$th percentile is the value $a$ with $n\%$ of the distribution below it. The median is the 50th percentile. You are looking for a bound this time. You know the area, so you solve for the bound. So you solve $F(a)=\frac{n}{100}$.
For the PDF $f(x)=\frac{3}{2}\left(1-x^{2}\right)$ on $0\le x\le1$, the 47th percentile is the value of $a$ that solves
On Paper 2, use your calculator's solver. Be warned that the Casio fx-CG50 can take one to three minutes to solve an equation with an integral in it. Set it going, then start reading the next question while you wait.
Game plan: Percentiles
1Write the integral from the bottom of the interval up to $a$, and set it equal to the percentage as a decimal.
2Integrate and solve for $a$, by hand or with a solver.
3Reject any solution that is outside the interval where the PDF is defined.
Watch the area split
Practise
Exam question: a percentile
2022 Paper 1, Question 18
Complex unfamiliar · Technology-free · [4 marks]
A percentile is a measure in statistics showing the value below which a given percentage of observations occur.
The continuous random variable $X$ has the probability density function
Reject $a=0.4$, because it is outside $1\le x\le2$.
1 markJudging which value is reasonable, using the domain.
Answer
The 36th percentile is 1.6.
Practice
Practice questions
Eight questions of my own, covering every skill. Each one is labelled with the skill it practises, so if one goes wrong you know which part of the guide to go back to.
Question 1 of 8
Skill 1: estimates from data · Multiple choice · Technology-free · [1 mark]
A company tested 200 phone batteries. 46 of them lasted between 20 and 25 hours. An estimate of the probability that a battery of this type lasts between 20 and 25 hours is
The relative frequency is the number in the interval divided by the total: $\frac{46}{200}=0.23$. On a histogram, this is the area of the bar for that interval.
Option (C) has divided by 100 instead of 200, and (D) is the probability of lasting outside that interval.
Answer
(B) 0.23
Question 2 of 8
Skill 2: probability as area · Technology-free · [3 marks]
The continuous random variable $X$ has the probability density function shown. Determine the value of $k$.
Skill 2: probability as area · Technology-free · [3 marks]
The cumulative distribution function of a continuous random variable $X$ is $F(x)=\dfrac{x^{2}}{16}$ for $0\le x\le4$. Determine $P(1\le X\le3)$, and the probability density function of $X$.
1$F(x)$ is the area up to $x$, so the area between 1 and 3 is $F(3)-F(1)$.
Option (B) is the middle of the interval, which would only be right if the density were flat. Here the density is bigger near 1, so the mean is pulled above $\frac12$.
Answer
(C) $\frac{2}{3}$
Question 7 of 8
Skill 2: probability as area · Technology-free · [3 marks]
The continuous random variable $X$ has the probability density function shown. Determine the median of $X$.
3Reject $-3\sqrt2$, because it is outside $0\le x\le6$.
Answer
$m=3\sqrt2$
Question 8 of 8
Skill 4: variance and standard deviation · Technology-active · [4 marks]
The continuous random variable $X$ has the probability density function shown. Determine the standard deviation of $X$, and the probability that $X$ is within one standard deviation of the mean.
$\sigma\approx0.447$, and the probability is about 0.626
Keep practising
Ready for something harder?
The percentile question has its own complex unfamiliar walkthrough. Try it again from a blank page first. The hard part is realising you are solving for a bound rather than working out an area.
1Unknown constant in the PDF? Set the total area equal to 1 and solve for it before anything else.
2Translate the question into a statement about $X$, especially if the variable has been shifted or scaled.
3Probability: integrate $f(x)$ between the bounds. Mean: integrate $x\,f(x)$. Variance: $E\left(X^{2}\right)-\big[E(X)\big]^{2}$. Standard deviation: square root the variance.
4Percentile or median: you know the area, so solve for the upper bound. Reject answers outside the interval.
5Sanity check: probabilities are between 0 and 1, and the mean and percentiles sit inside the interval.